As large language models (LLMs) evolve into autonomous agents, evaluating Repository-Level Reasoning, the ability to maintain logical consistency across massive, interdependent file systems, has become critical.Current benchmarks typically fluctuate between isolated code snippets and black-box evaluations.We present REPOREASON, a white-box diagnostic benchmark centered on Abductive Assertion Verification.To eliminate memorization while preserving authentic logical depth, we implement an Execution-Driven Mutation framework that utilizes the environment as a Semantic Oracle to regenerate ground-truth states.Furthermore, we establish a fine-grained diagnostic system using dynamic program slicing, quantifying reasoning via three orthogonal metrics: ESV (Reading Load), MCL (Simulation Depth), and DFI (Integration Width).Comprehensive evaluations of frontier models (e.g., Claude-4.5-Sonnet,DeepSeek-v3.1-Terminus)reveal a prevalent Aggregation Deficit, where integration width serves as the primary cognitive bottleneck.Our findings provide granular whitebox insights for optimizing the next generation of agentic software engineering.

Software logging is critical for system observability, yet developers face a dual crisis of costly overlogging and risky underlogging. Existing automated logging tools often overlook the fundamental whether-to-log decision and struggle with the composite nature of logging. In this paper, we propose AutoLogger, a novel hybrid framework that addresses the complete the end-to-end logging pipeline. AutoLogger first employs a fine-tuned classifier, the Judger, to accurately determine if a method requires new logging statements. If logging is needed, a multi-agent system is activated. The system includes specialized agents: a Locator dedicated to determining where to log, and a Generator focused on what to log. These agents work together, utilizing our designed program analysis and retrieval tools. We evaluate AutoLogger on a large corpus from three mature open-source projects against state-of-the-art baselines. Our results show that AutoLogger achieves 96.63% F1-score on the crucial whether-to-log decision. In an end-to-end setting, AutoLogger improves the overall quality of generated logging statements by 16.13% over the strongest baseline, as measured by an LLM-as-a-judge score. We also demonstrate that our framework is generalizable, consistently boosting the performance of various backbone LLMs.

Code retrieval aims to provide users with desired code snippets based on users’ natural language queries. With the development of deep learning technologies, adopting pre-trained models for this task has become mainstream. Considering the retrieval efficiency, most of the previous approaches adopt a dual-encoder for this task, which encodes the description and code snippet into representation vectors, respectively. However, the model structure of the dual-encoder tends to limit the model’s performance, since it lacks the interaction between the code snippet and description at the bottom layer of the model during training. To improve the model’s effectiveness while preserving its efficiency, we propose a framework, which adopts self-adaptive model distillation for efficient code retrieval (SPENCER). SPENCER first adopts the dual-encoder to narrow the search space and then adopts the cross-encoder to improve accuracy. To improve the efficiency of SPENCER, we propose a novel model distillation technique, which can greatly reduce the inference time of the dual-encoder while maintaining the overall performance. We also propose a teaching assistant selection strategy for our model distillation, which can adaptively select the suitable teaching assistant models for different pre-trained models during the model distillation to ensure the model performance. Extensive experiments demonstrate that the combination of dual-encoder and cross-encoder improves overall performance compared to solely dual-encoder-based models for code retrieval. Besides, our model distillation technique retains over 98% of the overall performance while reducing the inference time of the dual-encoder by 70%.

The rapid advancement of AI has led to its integration into various areas, especially with Large Language Models (LLMs) significantly enhancing capabilities in Artificial Intelligence Generated Content (AIGC). However, the complexity of AI systems has also exposed their vulnerabilities, necessitating robust methods for Failure Analysis (FA) and Fault Injection (FI) to ensure resilience and reliability. Despite the importance of these techniques, there lacks a comprehensive review of FA and FI methodologies in AI systems. This study fills this gap by presenting a detailed survey of existing FA and FI approaches across six layers of AI systems. We systematically analyze 142 studies to answer three research questions including (1) what are the prevalent failures in AI systems, (2) what types of faults can current FI tools simulate, (3) what gaps exist between the simulated faults and real-world failures. Our findings reveal a taxonomy of AI system failures, assess the capabilities of existing FI tools, and highlight discrepancies between real-world and simulated failures. Moreover, this survey contributes to the field by providing a framework for fault diagnosis, evaluating the state-of-the-art in FI, and identifying areas for improvement in FI techniques to enhance the resilience of AI systems.

Developers write logging statements to monitor software runtime behaviors and system state. However, poorly constructed or misleading log messages can inadvertently obfuscate actual program execution patterns, thereby impeding effective software maintenance. Existing research on analyzing issues within logging statements is limited, primarily focusing on detecting a singular type of defect and relying on manual intervention for fixes rather than automated solutions. To address the limitation, we initiate a systematic study that pinpoints four specific types of defects in logging statements (i.e., statement code inconsistency, static dynamic inconsistency, temporal relation inconsistency, and readability issues) through the analysis of real-world log-centric changes. We then propose LogUpdater , a two-stage framework for automatically detecting and updating logging statements for these specific defects. In the offline stage, LogUpdater constructs a similarity-based classifier on a set of synthetic defective logging statements to identify specific defect types. During the online testing phase, this classifier first evaluates logging statements in a given code snippet to determine the necessity and type of improvements required. Then, LogUpdater constructs type-aware prompts from historical logging update changes for an LLM-based recommendation framework to suggest updates addressing these specific defects. We evaluate the effectiveness of LogUpdater on a dataset containing real-world logging changes, a synthetic dataset, and a new real-world project dataset. The results indicate that our approach is highly effective in detecting logging defects, achieving an F1-score of 0.625. Additionally, it exhibits significant improvements in suggesting precise static text and dynamic variables, with enhancements of 48.12% and 24.90%, respectively. Furthermore, LogUpdater achieves a 61.49% success rate in recommending correct updates on new real-world projects. We reported 40 problematic logging statements and their fixes to GitHub via pull requests, resulting in 25 changes confirmed and merged across 11 different projects.

Solving complex tasks in a single attempt is challenging for large language models (LLMs). Iterative interaction with the environment and feedback is often required to achieve success, making effective feedback utilization a critical topic. Existing approaches either struggle with length generalization or rely on naive retries without leveraging prior information. In this paper, we introduce FTTT, a novel paradigm that formulates feedback utilization as an optimization problem at test time. Additionally, we propose a learnable test-time optimizer, OpTune, to effectively exploit feedback. Experiments on two LLMs across four reasoning datasets demonstrate that FTTT and OpTune achieve superior scalability and performance.

The significant breakthroughs of Medical Multi-Modal Large Language Models (Med-MLLMs) renovate modern healthcare with robust information synthesis and medical decision support. However, these models are often evaluated on benchmarks that are unsuitable for the Med-MLLMs due to the complexity of real-world diagnostics across diverse specialties. To address this gap, we introduce Asclepius, a novel Med-MLLM benchmark that comprehensively assesses Med-MLLMs in terms of: distinct medical specialties (cardiovascular, gas-troenterology, etc.) and different diagnostic capacities (perception, disease analysis, etc.). Grounded in 3 proposed core principles, Asclepius ensures a comprehensive evaluation by encompassing 15 medical specialties, stratifying into 3 main categories and 8 sub-categories of clinical tasks, and exempting overlap with existing VQA dataset. We further provide an in-depth analysis of 6 Med-MLLMs and compare them with 3 human specialists, providing insights into their competencies and limitations in various medical contexts. Our work not only advances the understanding of Med-MLLMs' capabilities but also sets a precedent for future evaluations and the safe deployment of these models in clinical environments. © 2025 Association for Computational Linguistics.

Rust has been extensively used in software and system development due to its guarantees for memory and concurrency safety. Fuzzing is a popular bug detection technique for examining the correctness and robustness of programs. However, we identify that current state-of-the-art Rust fuzzers are significantly impeded by the ubiquitous presence of Rust runtime safety checks, resulting in poor effectiveness and efficiency. These checks, which are inserted either implicitly by the compiler or explicitly by the compiler or developers, could cause a high number of panic crashes and early program termination in fuzzing. Consequently, current fuzzers are unable to effectively explore deep code behind the runtime safety checks, leaving potential vulnerabilities undetected.

Cheryl Lee, Chunqiu Steven Xia, Longji Yang, Jen-tse Huang, Zhouruixing Zhu, Lingming Zhang, Michael R. Lyu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Wenxin Tang, Jingyu Xiao, Wenxuan Jiang, Xi Xiao, Yuhang Wang, Xuxin Tang, Qing Li, Yuehe Ma, Junliang Liu, Shisong Tang, Michael R. Lyu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Wenxuan Wang, Shi Juluan, Zixuan Ling, Yuk-Kit Chan, Chaozheng Wang, Cheryl Lee, Youliang Yuan, Jen-tse Huang, Wenxiang Jiao, Michael R. Lyu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Background: It has long been suggested that user feedback, typically written in natural language by end-users, can help issue detection. However, for large-scale online service systems that receive a tremendous amount of feedback, it remains a challenging task to identify severe issues from user feedback. Aims: To develop a better feedback-based issue detection approach, it is crucial first to gain a comprehensive understanding of the characteristics of user feedback in real production systems. Method: In this paper, we conduct an empirical study on 50,378,766 user feedback items from six real-world services in a one-billion-user online service system. We first study what users provide in their feedback. We then examine whether certain features of feedback items can be good indicators of severe issues. Finally, we investigate whether adopting machine learning techniques to analyze user feedback is reasonable. Results: Our results show that a large proportion of user feedback provides irrelevant information about system issues. As a result, it is crucial to filter out issue-irrelevant information when processing user feedback. Moreover, we find severe issues that cannot be easily detected based solely on user feedback characteristics. Finally, we find that the distributions of the feedback topics in different time intervals are similar. This confirms that designing machine learning-based approaches is a viable direction for better analyzing user feedback. Conclusions: We consider that our findings can serve as an empirical foundation for feedback-based issue detection in large-scale service systems, which sheds light on the design and implementation of practical issue detection approaches.

Decision-making is a complex process requiring diverse abilities, making it an excellent framework for evaluating Large Language Models (LLMs). Researchers have examined LLMs' decision-making through the lens of Game Theory. However, existing evaluation mainly focus on two-player scenarios where an LLM competes against another. Additionally, previous benchmarks suffer from test set leakage due to their static design. We introduce GAMA($\gamma$)-Bench, a new framework for evaluating LLMs' Gaming Ability in Multi-Agent environments. It includes eight classical game theory scenarios and a dynamic scoring scheme specially designed to quantitatively assess LLMs' performance. $\gamma$-Bench allows flexible game settings and adapts the scoring system to different game parameters, enabling comprehensive evaluation of robustness, generalizability, and strategies for improvement. Our results indicate that GPT-3.5 demonstrates strong robustness but limited generalizability, which can be enhanced using methods like Chain-of-Thought. We also evaluate 13 LLMs from 6 model families, including GPT-3.5, GPT-4, Gemini, LLaMA-3.1, Mixtral, and Qwen-2. Gemini-1.5-Pro outperforms others, scoring of $69.8$ out of $100$, followed by LLaMA-3.1-70B ($65.9$) and Mixtral-8x22B ($62.4$). Our code and experimental results are publicly available at https://github.com/CUHK-ARISE/GAMABench.

Large language model-based multi-agent systems have shown great abilities across various tasks due to the collaboration of expert agents, each focusing on a specific domain. However, the impact of clumsy or even malicious agents—those who frequently make errors in their tasks—on the overall performance of the system remains underexplored. This paper investigates: (1) What is the resilience of various system structures (e.g., A$\rightarrow$B$\rightarrow$C, A$\leftrightarrow$B$\leftrightarrow$C) under faulty agents, on different downstream tasks? (2) How can we increase system resilience to defend against these agents? To simulate faulty agents, we propose two approaches—AutoTransform and AutoInject—which introduce mistakes into the agents’ responses. Experiments on four downstream tasks using six systems show that the "hierarchical" structure, i.e., A$\rightarrow$(B$\leftrightarrow$C), exhibits superior resilience with the lowest performance drop of 5.5%, compared to 10.5% and 23.7% of other two structures. To further improve resilience, we introduce (1) Challenger, that introduces a mechanism for each agent to challenge others’ outputs, and (2) Inspector, an additional agent to review and correct messages, recovering up to 96.4% errors made by faulty agents. Our code and data are available at https://github.com/CUHK-ARISE/MAS-Resilience.

Runtime failures are commonplace in modern distributed systems. When such issues arise, users often turn to platforms such as Github or JIRA to report them and request assistance. Automatically identifying the root cause of these failures is critical for ensuring high reliability and availability. However, prevailing automatic root cause analysis (RCA) approaches rely significantly on comprehensive runtime monitoring data, which is often not fully available in issue platforms. Recent methods leverage large language models (LLMs) to analyze issue reports, but their effectiveness is limited by incomplete or ambiguous user-provided information. To obtain more accurate and comprehensive RCA results, the core idea of this work is to extract additional diagnostic clues from code to supplement data-limited issue reports. Specifically, we propose COCA, a code knowledge enhanced root cause analysis approach for issue reports. Based on the data within issue reports, COCA intelligently extracts relevant code snippets and reconstructs execution paths, providing a comprehensive execution context for further RCA. Subsequently, COCA constructs a prompt combining historical issue reports along with profiled code knowledge, enabling the LLMs to generate detailed root cause summaries and localize responsible components. Our evaluation on datasets from five real-world distributed systems demonstrates that COCA significantly outperforms existing methods, achieving a$\mathbf{2 8. 3 \%}$improvement in root cause localization and a 22.0 % improvement in root cause summarization. Furthermore, COCA's performance consistency across various LLMs underscores its robust generalizability.

The code written by developers usually suffers from efficiency problems and contain various performance bugs. These inefficiencies necessitate the research of automated refactoring methods for code optimization. Early research in code optimization employs rule-based methods and focuses on specific inefficiency issues, which are labor-intensive and suffer from the low coverage issue. Recent work regards the task as a sequence generation problem, and resorts to deep learning (DL) techniques such as large language models (LLMs). These methods typically prompt LLMs to directly generate optimized code. Although these methods show state-of-the-art performance, such one-step generation paradigm is hard to achieve an optimal solution. First, complex optimization methods such as combinatorial ones are hard to be captured by LLMs. Second, the one-step generation paradigm poses challenge in precisely infusing the knowledge required for effective code optimization within LLMs, resulting in under-optimized code. To address these problems, we propose to model this task from the search perspective, and propose a search-based LLMs framework named SBLLM that enables iterative refinement and discovery of improved optimization methods. SBLLM synergistically integrate LLMs with evolutionary search and consists of three key components: 1) an execution-based representative sample selection part that evaluates the fitness of each existing optimized code and prioritizes promising ones to pilot the generation of improved code; 2) an adaptive optimization pattern retrieval part that infuses targeted optimization patterns into the model for guiding LLMs towards rectifying and progressively enhancing their optimization methods; and 3) a genetic operatorinspired chain-of-thought prompting part that aids LLMs in combining different optimization methods and generating improved optimization methods. Our evaluation of SBLLM on a dataset of Python and C++ code demonstrates its effectiveness in improving code efficiency. Specifically, the results indicate that SBLLM can improve program execution efficiency by up to 209.59 % and consistently outperform all baseline methods by$8.75 \% \sim 28.06 {\%}$and$1.15 \% \sim 9.56 {\%}$with different LLMs in terms of top-5 speedup rate on Python and C++, respectively.

A common practice in the reliability engineering of cloud services involves the collection of monitoring metrics, followed by comprehensive analysis to identify performance issues. However, existing methods often fall short of detecting diverse and evolving anomalies across different services. More-over, there exists a significant gap between the technical and business interpretation of anomalies, i.e., a detected anomaly may not have an actual impact on system performance or user experience. To address these challenges, we propose ADAMAS, an adaptive AutoML-based anomaly detection framework aiming to achieve practical anomaly detection in production cloud systems. To improve the ability to detect cross-service anomalies, we design a novel unsupervised evaluation function to facilitate the automatic searching of the optimal model structure and parameters. ADAMAS also contains a lightweight human-in-the-loop design, which can efficiently incorporate expert knowledge to adapt to the evolving anomaly patterns and bridge the gap between predicted anomalies and actual business exceptions. Fur-thermore, through monitoring the rate of mispredicted anomalies, ADAMAS proactively re-configures the optimal model, forming a continuous loop of system improvement. Extensive evaluation on one public and two industrial datasets shows that ADAMAS outperforms all baseline models with a 0.891 F1-score. The ablation study also proves the effectiveness of the evaluation function design and the incorporation of expert knowledge.

Code retrieval, which retrieves code snippets based on users' natural language descriptions, is widely used by devel-opers and plays a pivotal role in real-world software development. The advent of deep learning has shifted the retrieval paradigm from lexical-based matching towards leveraging deep learning models to encode source code and queries into vector represen-tations, facilitating code retrieval according to vector similarity. Despite the effectiveness of these models, managing large-scale code database presents significant challenges. Previous research proposes deep hashing-based methods, which generate hash codes for queries and code snippets and use Hamming distance for rapid recall of code candidates. However, this approach's reliance on linear scanning of the entire code base limits its scalability. To further improve the efficiency of large-scale code retrieval, we propose a novel approach SECRET (Scalable and Efficient Code Retrieval via SegmEnTed deep hashing). SECRET converts long hash codes calculated by existing deep hashing approaches into several short hash code segments through an iterative training strategy. After training, SECRET recalls code candidates by looking up the hash tables for each segment, the time complexity of recall can thus be greatly reduced. Extensive experimental results demonstrate that SECRET can drastically reduce the retrieval time by at least 95 % while achieving comparable or even higher performance of existing deep hashing approaches. Besides, SECRET also exhibits superior performance and efficiency compared to the classical hash table-based approach known as LSH under the same number of hash tables.

Automated translation of legacy$\mathbf{C}$code into Rust aims to ensure memory safety while reducing the burden of manual migration. Early approaches in C-to-Rust translation rely on static rule-based methods, but they suffer from limited coverage due to dependence on predefined rule patterns. Recent works regard the task as a sequence-to-sequence problem by leveraging large language models (LLMs). Although these LLM-based methods are capable of reducing unsafe code blocks, the translated code often exhibits issues in following Rust rules and maintaining semantic consistency. On one hand, existing methods adopt a direct prompting strategy to translate the$C$code, which struggles to accommodate the syntactic rules between C and Rust. On the other hand, this strategy makes it difficult for LLMs to accurately capture the semantics of complex code. To address these challenges, we propose IRENE, an LLM-based framework that Integrates RulEs aNd sEmantics to enhance translation. IRENE consists of three modules: 1) a ruleaugmented retrieval module that selects relevant translation examples based on rules generated from a static analyzer developed by us, thereby improving the handling of Rust rules; 2) a structured summarization module that produces a structured summary for guiding LLMs to enhance the semantic understanding of$\mathbf{C}$code; 3) an error-driven translation module that leverages compiler diagnostics to iteratively refine translations. We evaluate IRENE on two datasets (xCodeEval-a public dataset, HW-Bench-an industrial dataset provided by Huawei) and eight LLMs, focusing on translation accuracy and safety. In the xCodeEval, IRENE consistently outperforms the strongest baseline method in all LLMs, achieving average improvements of 8.06% and 12.74% in the computational accuracy (CA) and compilation success rate (CSR), respectively. It also enhances the safety of translated code, reducing the Unsafe Rate (UR) to$\mathbf{1. 7 0} \boldsymbol{\%}$on average. In the HW-Bench, when compared to the strongest baseline, IRENE improves CSR and reduces UR by an average of$\mathbf{0. 3 3 \%}$and$\mathbf{2 6. 0 0 \%}$, respectively.

Managing complex cloud services requires standard operational documentation, but its sheer volume often hinders cloud engineers from efficient knowledge acquisition. Retrieval-Augmented Generation (RAG) can streamline this process by retrieving relevant knowledge and generating concise, referenced answers. However, deploying a reliable RAG-based chatbot for cloud operation remains a challenge. In this experience paper, we analyze the development and deployment of RAG-based chatbots for operational question answering (OpsQA) at a large-scale cloud vendor. Through an empirical study of 2,000 real-world queries across three operational teams, we identify five unique OpsQA intent types (e.g., symptom analysis and terminology explanation) and their corresponding requirements for a satisfactory answer, which differ from general software engineering queries. Our analysis further uncovers six root causes leading to chatbot failures—over half stem from query issues (i.e., incompleteness, out-of-scope, or invalid queries), while others are from retrieval or generation issues. To address these issues, we propose iKnow, an intent-guided RAG-based chatbot that integrates intent detection, query rewriting tailored to each intent, and missing knowledge detection to enhance answer quality. In internal evaluations, iKnow improves average answer accuracy from 65.8% to 81.3% with only a modest increase in latency. iKnow has been deployed for six months at CloudA, supporting thousands of cloud engineers in daily operations. We discuss lessons learned from real-world deployment, providing valuable insights for future research and practical implementations in similar domains.

Effective alert diagnosis is essential for ensuring the reliability of large-scale online service systems. However, on-call engineers are often burdened with manually inspecting massive volumes of logs to identify root causes. While various automated tools have been proposed, they struggle in practice due to alert-agnostic log scoping and the inability to organize complex data effectively for reasoning. To overcome these limitations, we introduce LogPilot, an intent-aware and scalable framework powered by Large Language Models (LLMs) for automated log-based alert diagnosis. LogPilot introduces an intent-aware approach, interpreting the logic in alert definitions (e.g., PromQL) to precisely identify causally related logs and requests. To achieve scalability, it reconstructs each request’s execution into a spatiotemporal log chain, clusters similar chains to identify recurring execution patterns, and provides representative samples to the LLMs for diagnosis. This clustering-based approach ensures the input is both rich in diagnostic detail and compact enough to fit within the LLM’s context window. Evaluated on real-world alerts from Volcano Engine Cloud, LogPilot improves the usefulness of root cause summarization by 50.34% and exact localization accuracy by 54.79% over state-of-the-art methods. With a diagnosis time under one minute and a cost of only $0.074 per alert, LogPilot has been successfully deployed in production, offering an automated and practical solution for service alert diagnosis.

High-quality logging is critical for the reliability of cloud services, yet the industrial process for improving it is typically manual, reactive, and unscalable. Existing automated tools inherit this reactive nature, failing to answer the crucial whether-to-log question and are constrained to simple logging statement insertion, thus addressing only a fraction of the real-world logging improvement.To address these gaps and cope with logging debt in large-scale codebases, we propose LogImprover, a framework powered by Large Language Models (LLMs) that automates proactive logging quality improvement. LogImprover introduces two paradigm shifts: from reactive generation to proactive discovery, and from simple insertion to holistic logging patch generation. First, it identifies potential logging gaps based on principles distilled from industrial best practices. Then, it grounds each candidate through a cascading, structure-aware RAG module. Next, it prunes false positives by analyzing call-stack logging responsibilities and implicit logger inheritance. Finally, it generates holistic and explainable logging patches that reflect real-world development practices.Our evaluation provides dual confirmation of its effectiveness: LogImprover significantly outperforms state-of-the-art base-lines in closed-world experiments and achieves 68.12% developer acceptance rate in its real-world deployment. This success demonstrates the practical value of automating the entire logging quality improvement lifecycle, from discovery to recommendation.

The rapid growth of audio-centric platforms and applications such as Whatsapp and Twitter has transformed the way people communicate and share audio content in modern society. However, these platforms are increasingly misused to disseminate harmful audio content, such as hate speech, deceptive advertisements, and explicit material, which can have significant negative consequences (e.g., detrimental effects on mental health). In response, researchers and practitioners have been actively developing and deploying audio content moderation tools to tackle this issue. Despite these efforts, malicious actors can bypass moderation systems by making subtle alterations to audio content, such as modifying pitch or inserting noise. Moreover, the effectiveness of modern audio moderation tools against such adversarial inputs remains insufficiently studied. To address these challenges, we propose MTAM, a Metamorphic Testing framework for Audio content Moderation software. Specifically, we conduct a pilot study on 2000 audio clips and define 14 metamorphic relations across two perturbation categories: Audio Features-Based and Heuristic perturbations. MTAM applies these metamorphic relations to toxic audio content to generate test cases that remain harmful while being more likely to evade detection. In our evaluation, we employ MTAM to test five commercial textual content moderation software and an academic model against three kinds of toxic content. The results show that MTAM achieves up to 38.6%, 18.3%, 35.1%, 16.7%, and 51.1% error finding rates (EFR) when testing commercial moderation software provided by Gladia, Assembly AI, Baidu, Nextdata, and Tencent respectively, and it obtains up to 45.7% EFR when testing the state-of-the-art algorithms from the academy. In addition, we leverage the test cases generated by MTAM to retrain the model we explored, which largely improves model robustness (nearly 0% EFR) while maintaining the accuracy on the original test set. We release the code and experiment data to facilitate future research1.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable performance on the design-to-code task, i.e., generating UI code from UI mock-ups. However, existing benchmarks only contain static web pages for evaluation and ignore the dynamic interaction, limiting the practicality, usability and user engagement of the generated webpages.To bridge these gaps, we present the first systematic investigation of MLLMs in generating interactive webpages. Specifically, we formulate the Interaction-to-Code task and establish the Interaction2Code benchmark, encompassing 127 unique webpages and 374 distinct interactions across 15 webpage types and 31 interaction categories. Through comprehensive experiments utilizing state-of-the-art (SOTA) MLLMs, evaluated via both automatic metrics and human assessments, we identify four critical limitations of MLLM on Interaction-to-Code task: (1) inadequate generation of interaction compared with full page, (2) prone to ten types of failure, (3) poor performance on visually subtle interactions, and (4) insufficient undestanding on interaction when limited to single-modality visual descriptions. To address these limitations, we propose four enhancement strategies: Interactive Element Highlighting, Failure-aware Prompting (FAP), Visual Saliency Enhancement, and Visual-Textual Descriptions Combination, all aiming at improving MLLMs’ performance on the Interaction-to-Code task. Our data and code are available in https://github.com/WebPAI/Interaction2Code.

Large Language Models (LLMs) have recently demonstrated strong capabilities in code-related tasks, but their robustness in code reasoning under perturbations remains underexplored. We introduce CodeCrash, a stress-testing framework with 1,279 questions from CRUXEVAL and LIVECODEBENCH, designed to evaluate reasoning reliability under structural perturbations and misleading natural language (NL) contexts. Through a systematic evaluation of 17 LLMs, we find that models often shortcut reasoning by over-relying on NL cues, leading to an average performance degradation of 23.2% in output prediction tasks. Even with Chain-of-Thought reasoning, models on average still have a 13.8% drop due to distractibility and rationalization, revealing a lack of critical reasoning capability to distinguish the actual code behaviors. While Large Reasoning Models with internal reasoning mechanisms improve robustness by fostering critical thinking, plausible yet incorrect hints can trigger pathological self-reflection, causing 2-3 times token consumption and even catastrophic cognitive dissonance in extreme cases for QwQ-32B. We refer to this phenomenon as Reasoning Collapse. CodeCrash provides a rigorous benchmark for evaluating robustness in code reasoning, guiding future research and development toward more reliable and resilient models.

Clinical decision making (CDM) is a complex, dynamic process crucial to healthcare delivery, yet it remains a significant challenge for artificial intelligence systems. While Large Language Model (LLM)-based agents have been tested on general medical knowledge using licensing exams and knowledge question-answering tasks, their performance in the CDM in real-world scenarios is limited due to the lack of comprehensive benchmark that mirror actual medical practice. To address this gap, we present MedChain, a dataset of 12,163 clinical cases that covers five key stages of clinical workflow. MedChain distinguishes itself from existing benchmarks with three key features of real-world clinical practice: personalization, interactivity, and sequentiality. Further, to tackle real-world CDM challenges, we also propose MedChain-Agent, an AI system that integrates a feedback mechanism and a MedCase-RAG module to learn from previous cases and adapt its responses. MedChain-Agent demonstrates remarkable adaptability in gathering information dynamically and handling sequential clinical tasks, significantly outperforming existing approaches. The relevant dataset and code will be released upon acceptance of this paper.

WebAssembly (Wasm) is a binary instruction format designed as a portable compilation target, which has been widely used on both the web and server sides in recent years. As high performance is a critical design goal of Wasm, it is essential to conduct performance testing for Wasm runtimes. However, existing research on Wasm runtime performance testing still suffers from insufficient high-quality test programs. To solve this problem, we propose a novel test program generation approach WarpGen. It first extracts code snippets from historical issue-triggering test programs as initial operators, then inserts an operator into a seed program to synthesize a new test program. To verify the quality of generated programs, we propose an indicator called distinguishability, which refers to the ability of a test program to distinguish abnormal performance of specific Wasm runtimes. We apply WarpGen for performance testing on four Wasm runtimes and verify its effectiveness compared with baseline approaches. In particular, WarpGen has identified seven new performance issues in three Wasm runtimes.

Cloud systems, typically comprised of various components (e.g., microservices), are susceptible to performance issues, which may cause service-level agreement violations and financial losses. Identifying performance issues is thus of paramount importance for cloud vendors. In current practice, crucial metrics, i.e., Key Performance Indicators (KPIs), are monitored periodically to provide insight into the operational status of components. Identifying performance issues is often formulated as an anomaly detection problem, which is tackled by analyzing each metric independently. However, this approach overlooks the complex dependencies existing among cloud components. Some graph neural network-based methods take both temporal and relational information into account; however, the correlation violations in the metrics that serve as indicators of underlying performance issues are difficult for them to identify. Furthermore, a large volume of components in a cloud system results in a vast array of noisy metrics. This complexity renders it impractical for engineers to fully comprehend the correlations, making it challenging to identify performance issues accurately. To address these limitations, we propose Identifying Performance Issues based on Relational-Temporal Features (ISOLATE), a learning-based approach that leverages both the relational and temporal features of metrics to identify performance issues. In particular, it adopts a graph neural network with attention to characterizing the relations among metrics and extracts long-term and multi-scale temporal patterns using a GRU and a convolution network, respectively. The learned graph attention weights can be further used to localize the correlation-violated metrics. Moreover, to relieve the impact of noisy data, ISOLATE utilizes a Positive Unlabeled (PU) Learning strategy that tags pseudo-labels based on a small portion of confirmed negative examples. Extensive evaluation on both public and industrial datasets shows that ISOLATE outperforms all baseline models with 0.945 F1 score and 0.920 Hit rate@3. The ablation study also proves the effectiveness of the relational-temporal features and the PU-Learning strategy. Furthermore, we share the success stories of leveraging ISOLATE to identify performance issues in Huawei Cloud, which demonstrates its superiority in practice.

Automatic programming has seen increasing popularity due to the emergence of tools like GitHub Copilot which rely on Large Language Models (LLMs). At the same time, automatically generated code faces challenges during deployment due to concerns around quality and trust. In this article, we study automated coding in a general sense and study the concerns around code quality, security, and related issues of programmer responsibility. These are key issues for organizations while deciding on the usage of automatically generated code. We discuss how advances in software engineering such as program repair and analysis can enable automatic programming. We conclude with a forward looking view, focusing on the programming environment of the near future, where programmers may need to switch to different roles to fully utilize the power of automatic programming. Automated repair of automatically generated programs from LLMs can help produce higher assurance code from LLMs, along with evidence of assurance.

Transformer-based models have demonstrated state-of-the-art performance in various intelligent coding tasks such as code comment generation and code completion. Previous studies show that deep learning models are sensitive to input variations, but few have systematically studied the robustness of Transformer under perturbed input code. In this work, we empirically study the effect of semantic-preserving code transformations on the performance of Transformers. Specifically, 27 and 24 code transformation strategies are implemented for two popular programming languages, Java and Python, respectively. To facilitating analysis, the strategies are grouped into five categories: block transformation, insertion / deletion transformation, grammatical statement transformation, grammatical token transformation, and identifier transformation. Experiments on three popular code intelligence tasks, including code completion, code summarization, and code search, demonstrate that insertion / deletion transformation and identifier transformation have the greatest impact on the performance of Transformers. Our results also suggest that Transformers based on abstract syntax trees (ASTs) show more robust performance than models based only on code sequences under most code transformations. Besides, the design of positional encoding can impact the robustness of Transformers under code transformations. We also investigate substantial code transformations at the strategy level to expand our study and explore other factors influencing the robustness of Transformers. Furthermore, we explore applications of code transformations. Based on our findings, we distill insights about the challenges and opportunities for Transformer-based code intelligence from various perspectives.

editorial reviewed

Distributed tracing serves as a fundamental element in the monitoring of cloud-based and datacenter systems. It provides visibility into the full life cycle of a request or operation across multiple services, which is essential for understanding system dependencies and performance bottlenecks. To mitigate computational and storage overheads, most tracing frameworks adopt a uniform sampling strategy, which inevitably captures overlapping and redundant information. More advanced methods employ learning-based approaches to bias the sampling toward more informative traces. However, existing methods fall short of considering the high-dimensional and dynamic nature of trace data, which is essential for the production deployment of trace sampling. To address these practical challenges, in this paper we present Trace Mesh,a scalable and streaming sampler for distributed traces. Tracemesh employs Locality-Sensitivity Hashing (LSH) to improve sampling efficiency by projecting traces into a low-dimensional space while preserving their similarity. In this process, Tracemesh accommodates previously unseen trace features in a unified and streamlined way. Subsequently, Tracemesh samples traces through evolving clustering, which dynamically adjusts the sampling decision to avoid over-sampling of recurring traces. The proposed method is evaluated with trace data collected from both open-source microservice benchmarks and production service systems. Ex-perimental results demonstrate that Tracemesh outperforms state-of-the-art methods by a significant margin in both sampling accuracy and efficiency.

Imperceptibility is one of the crucial requirements for adversarial examples. Previous adversarial attacks on 3D point cloud recognition suffer from noticeable outliers, resulting in low imperceptibility. We think that the drawbacks can be alleviated via taking the local curvature of the point cloud into consideration. Existing approaches introduce the local geometry distance into the attack objective function. However, their definition of the local geometry distance neglects different perceptibility of distortions along different directions. In this paper, we aim to enhance the imperceptibility of adversarial attacks on 3D point cloud recognition by better preserving the local curvature of the original 3D point clouds. To this end, we propose the Curvature-Invariant Method (CIM), which directly regularizes the back-propagated gradient during the generation of adversarial point clouds based on two assumptions. Specifically, we first decompose the back-propagated gradients into the tangent plane and the normal direction. Then we directly reduce the gradient along the large curvature direction on the tangent plane and only keep the gradient along the negative normal direction. Comprehensive experimental comparisons confirm the superiority of our approach. Notably, our strategy can achieve 7.2% and 14.5% improvements in Hausdorff distance and Gaussian curvature measurements of the imperceptibility.

With the great achievement of vision transformers (ViTs), transformer-based approaches have become the new paradigm for solving various computer vision tasks. However, recent research shows that similar to convolutional neural networks (CNNs), ViTs are still vulnerable to adversarial attacks. To explore the shared deficiency of models with different structures, researchers begin to analyze the cross-structure adversarial transferability, which is still under-explored. Therefore, in this work, we focus on the ViT attacks to improve the cross-structure transferability between the transformer-based and convolution-based models. Previous studies fail to thoroughly investigate the influence of the components inside the ViT models on adversarial transferability, leading to inferior performance. To overcome the drawback, we launch a motivating study by linearly down-scaling the gradients of components inside the ViT models to analyze their influence on adversarial transferability. Based on the motivating study, we find that the gradient of the skip connection most influences transferability and believe that back-propagating gradients from deeper blocks can enhance transferability. Therefore, we propose the Virtual Dense Connection method (VDC). Specifically, without changing the forward pass, we first recompose the original network to add virtual dense connections. Then we back-propagate gradients of deeper Attention maps and Multi-layer Perceptron (MLP) blocks via virtual dense connections when generating adversarial samples. Extensive experiments confirm the superiority of our proposed method over the state-of-the-art baselines, with an 8.2% improvement in transferability between ViT models and a 7.2% improvement in cross-structure transferability from ViTs to CNNs.

Wenxuan Wang, Wenxiang Jiao, Jingyuan Huang, Ruyi Dai, Jen-tse Huang, Zhaopeng Tu, Michael Lyu. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Recent advancements in long-context modeling have enhanced language models (LMs) for complex tasks across multiple NLP applications.Despite this progress, we find that these models struggle with multi-hop reasoning and exhibit decreased performance in the presence of noisy contexts.In this paper, we introduce Reasoning with Attributions, a novel approach that prompts LMs to supply attributions for each assertion during their reasoning.We validate our approach through experiments on three multi-hop datasets, employing both proprietary and open-source models, and demonstrate its efficacy and resilience.Furthermore, we explore methods to augment reasoning capabilities via fine-tuning and offer an attribution-annotated dataset and a specialized training strategy.Our fine-tuned model achieves competitive performance on multi-hop reasoning benchmarks, closely paralleling proprietary LMs such as ChatGPT and Claude-instant 1 .

Recent research has focused on examining Large Language Models' (LLMs) characteristics from a psychological standpoint, acknowledging the necessity of understanding their behavioral characteristics.The administration of personality tests to LLMs has emerged as a noteworthy area in this context.However, the suitability of employing psychological scales, initially devised for humans, on LLMs is a matter of ongoing debate.Our study aims to determine the reliability of applying personality assessments to LLMs, explicitly investigating whether LLMs demonstrate consistent personality traits.Analysis of 2,500 settings per model, including GPT-3.5, GPT-4, Gemini-Pro, and LLaMA-3.1, reveals that various LLMs show consistency in responses to the Big Five Inventory, indicating a satisfactory level of reliability.Furthermore, our research explores the potential of GPT-3.5 to emulate diverse personalities and represent various groups-a capability increasingly sought after in social sciences for substituting human participants with LLMs to reduce costs.Our findings reveal that LLMs have the potential to represent different personalities with specific prompt instructions.

We introduce LogicAsker, a novel approach for evaluating and enhancing the logical reasoning capabilities of large language models (LLMs) such as ChatGPT and GPT-4.Despite LLMs' prowess in tasks like writing assistance, code generation, and machine translation, assessing their ability to reason has been challenging.Traditional evaluations often prioritize accuracy on downstream tasks over direct assessments of reasoning processes.LogicAsker addresses this gap by employing a set of atomic reasoning skills grounded in propositional and predicate logic to systematically examine and improve the reasoning prowess of LLMs.Our methodology reveals significant gaps in LLMs' learning of logical rules, with identified reasoning failures ranging from 29% to 90% across different models.Moreover, we leverage these findings to construct targeted demonstration examples and fine-tune data, notably enhancing logical reasoning in models like GPT-4o by up to 5%.To our knowledge, this is the first effort to utilize test case outcomes to effectively refine LLMs' formal reasoning capabilities.We make our code, data, and results publicly available 1 to facilitate further research and replication of our findings.

Visual representation learning has been a cornerstone in computer vision, involving typical forms such as visual embeddings, structural symbols, and text-based representations.Despite the success of CLIP-type visual embeddings, they often lack access to world knowledge critical for visual reasoning.In this work, we propose Visual Table, a novel form of visual representation tailored for visual reasoning.Visual tables are constructed as hierarchical descriptions of visual scenes, featuring a scene description and multiple object-centric descriptions covering categories, attributes, and knowledge.Thanks to the structural and textual formats, visual tables offer unique properties over mere visual embeddings, such as explainability and controllable editing.Furthermore, they deliver instance-level world knowledge and detailed attributes that are essential for visual reasoning.To create visual tables, we develop a generator trained on the dataset with collected, small-scale annotations.Extensive results on 11 visual reasoning benchmarks demonstrate that the generated visual tables significantly outperform previous structural and text-based representations.Moreover, they consistently enhance state-of-theart multi-modal large language models across diverse benchmarks, showcasing their potential for advancing visual reasoning tasks.Our code is available at https://github.com/ LaVi-Lab/Visual-Table.

As an emerging cloud computing deployment paradigm, serverless computing is gaining traction due to its efficiency and ability to harness on-demand cloud resources. However, a significant hurdle remains in the form of the cold start problem, causing latency when launching new function instances from scratch. Existing solutions tend to use over-simplistic strategies for function pre-loading/unloading without full invocation pattern exploitation, rendering unsatisfactory optimization of the trade-off between cold start latency and resource waste. To bridge this gap, we propose SPES, the first differentiated scheduler for runtime cold start mitigation by optimizing serverless function provision. Our insight is that the common architecture of serverless systems prompts the concentration of certain invocation patterns, leading to predictable invocation behaviors. This allows us to categorize functions and pre-load/unload proper function instances with finer-grained strategies based on accurate invocation prediction. Experiments demonstrate the success of SPES in optimizing serverless function provision on both sides: reducing the 75th-percentile cold start rates by 49.77% and the wasted memory time by 56.43%, compared to the state-of-the-art. By mitigating the cold start issue, SPES is a promising advancement in facilitating cloud services deployed on serverless architectures.

Large Language Models (LLMs) have recently showcased their remarkable capacities, not only in natural language processing tasks but also across diverse domains such as clinical medicine, legal consultation, and education. LLMs become more than mere applications, evolving into assistants capable of addressing diverse user requests. This narrows the distinction between human beings and artificial intelligence agents, raising intriguing questions regarding the potential manifestation of personalities, temperaments, and emotions within LLMs. In this paper, we propose a framework, PsychoBench, for evaluating diverse psychological aspects of LLMs. Comprising thirteen scales commonly used in clinical psychology, PsychoBench further classifies these scales into four distinct categories: personality traits, interpersonal relationships, motivational tests, and emotional abilities. Our study examines five popular models, namely text-davinci-003, ChatGPT, GPT-4, LLaMA-2-7b, and LLaMA-2-13b. Additionally, we employ a jailbreak approach to bypass the safety alignment protocols and test the intrinsic natures of LLMs. We have made PsychoBench openly accessible via https://github.com/CUHK-ARISE/PsychoBench.

Pre-trained code models have recently achieved substantial improvements in many code intelligence tasks. These models are first pre-trained on large-scale unlabeled datasets in a task-agnostic manner using self-supervised learning, and then fine-tuned on labeled datasets in downstream tasks. However, the labeled datasets are usually limited in size (i.e., human intensive efforts), which may hinder the performance of pre-trained code models in specific tasks. To mitigate this, one possible solution is to leverage the large-scale unlabeled data in the tuning stage by pseudo-labeling, i.e., generating pseudo labels for unlabeled data and further training the pre-trained code models with the pseudo-labeled data. However, directly employing the pseudo-labeled data can bring a large amount of noise, i.e., incorrect labels, leading to suboptimal performance. How to effectively leverage the noisy pseudo-labeled data is a challenging yet under-explored problem.

As a dynamic programming language, Python has become increasingly popular in recent years. Although the dynamic type system of Python facilitates the developers in writing Python programs, it also brings type errors at run-time which are prevalent yet not easy to fix. There exist rule-based approaches for automatically repairing Python type errors. The approaches can generate accurate patches for the type errors covered by manually defined templates, but they require domain experts to design patch synthesis rules and suffer from low template coverage of real-world type errors. Learning-based approaches alleviate the manual efforts in designing patch synthesis rules and have become prevalent due to the recent advances in deep learning. Among the learning-based approaches, the prompt-based approach which leverages the knowledge base of code pre-trained models via pre-defined prompts, obtains state-of-the-art performance in general program repair tasks. However, such prompts are manually defined and do not involve any specific clues for repairing Python type errors, resulting in limited effectiveness. How to automatically improve prompts with the domain knowledge for type error repair is challenging yet under-explored.

Python is the top popular programming language used in the open-source community, largely owing to the extensive support from diverse third-party libraries within the PyPI ecosystem. Nevertheless, the utilization of third-party libraries can potentially lead to conflicts in dependencies, prompting researchers to develop dependency conflict detectors. Moreover, endeavors have been made to automatically infer dependencies. These approaches focus on version-level checks and inference, based on the assumption that configurations of libraries in the PyPI ecosystem are correct. However, our study reveals that this assumption is not universally valid, and relying solely on version-level checks proves inadequate in ensuring compatible run-time environments.

Unmanned aerial vehicles (UAVs) are becoming increasingly ubiquitous in our daily lives. However, like many other complex systems, UAVs are susceptible to software bugs that can lead to abnormal system behaviors and undesirable consequences. It is crucial to study such software bug-induced UAV anomalies, which are often manifested in flight logs, to help assure the quality and safety of UAV systems. However, there has been limited research on investigating the code-level patterns of software bug-induced UAV anomalies. This impedes the development of effective tools for diagnosing and localizing bugs within UAV system code.

Logs are imperative in the maintenance of online service systems, which often encompass important information for effective failure mitigation. While existing anomaly detection methodologies facilitate the identification of anomalous logs within extensive runtime data, manual investigation of log messages by engineers remains essential to comprehend faults, which is labor-intensive and error-prone. Upon examining the log-based troubleshooting practices at CloudA1, we find that engineers typically prioritize two categories of log information for diagnosis. These include fault-indicating descriptions, which record abnormal system events, and fault-indicating parameters, which specify the associated entities. Motivated by this finding, we propose an approach to automatically extract such fault-indicating information from logs for fault diagnosis, named LoFI. LoFI comprises two key stages. In the first stage, LoFI performs coarse-grained filtering to collect logs related to the faults based on semantic similarity. In the second stage, LoFI leverages a pre-trained language model with a novel prompt-based tuning method to extract fine-grained information of interest from the collected logs. We evaluate LoFI on logs collected from Apache Spark and an industrial dataset from CloudA. The experimental results demonstrate that LoFI outperforms all baseline methods by a significant margin, achieving an absolute improvement of 25.8˜37.9 in F1 over the best baseline method, ChatGPT. This highlights the effectiveness of LoFI in recognizing fault-indicating information. Furthermore, the successful deployment of LoFI at CloudA and user studies validate the utility of our method2.

To ensure the reliability of cloud systems, their run-time status reflecting the service quality is periodically monitored with monitoring metrics, i.e., KPIs (key performance indicators). When performance issues happen, root cause localization pinpoints the specific KPIs that are responsible for the degradation of overall service quality, facilitating prompt problem diagnosis and resolution. To this end, existing methods generally locate root-cause KPIs by identifying the KPIs that exhibit a similar anomalous trend to the overall service performance. While straightforward, solely relying on the similarity calculation may be ineffective when dealing with cloud systems with complicated interdependent services. Recent deep learning-based methods offer improved performance by modeling these intricate dependencies. However, their high computational demand often hinders their ability to meet the efficiency requirements of industrial applications. Furthermore, their lack of interpretability further restricts their practicality. To overcome these limitations, we propose KPIRoot, an effective and efficient method for root cause localization integrating both advantages of similarity analysis and causality analysis, where similarity measures the trend alignment of KPI and causality measures the sequential order of variation of KPI. Furthermore, we leverage symbolic aggregate approximation to produce a more compact representation for each KPI, enhancing the overall analysis efficiency of the approach. The experimental results show that KPIRoot outperforms seven state-of-the-art baselines by 7.9%~28.3%, while time cost is reduced by 56.9%. Moreover, we share our experience of deploying KPIRoot in the production environment of a large-scale cloud provider Cloud ${\mathcal{H}^{\ast}}$.

Log data have facilitated various tasks of software development and maintenance, such as testing, debugging and diagnosing. Due to the unstructured nature of logs, log parsing is typically required to transform log messages into structured data for automated log analysis. Given the abundance of log parsers that employ various techniques, evaluating these tools to comprehend their characteristics and performance becomes imperative. Loghub serves as a commonly used dataset for benchmarking log parsers, but it suffers from limited scale and representativeness, posing significant challenges for studies to comprehensively evaluate existing log parsers or develop new methods. This limitation is particularly pronounced when assessing these log parsers for production use. To address these limitations, we provide a new collection of annotated log datasets, denoted Loghub-2.0, which can better reflect the characteristics of log data in real-world software systems. Loghub-2.0 comprises 14 datasets with an average of 3.6 million log lines in each dataset. Based on Loghub-2.0, we conduct a thorough re-evaluation of 15 state-of-the-art log parsers in a more rigorous and practical setting. Particularly, we introduce a new evaluation metric to mitigate the sensitivity of existing metrics to imbalanced data distributions. We are also the first to investigate the granular performance of log parsers on logs that represent rare system events, offering in-depth details for software diagnosis. Accurately parsing such logs is essential, yet it remains a challenge. We believe this work could shed light on the evaluation and design of log parsers in practical settings, thereby facilitating their deployment in production systems.

Recently, there has been a growing interest in automatic software vulnerability detection. Pre-trained model-based approaches have demonstrated superior performance than other Deep Learning (DL)-based approaches in detecting vulnerabilities. However, the existing pre-trained model-based approaches generally employ code sequences as input during prediction, and may ignore vulnerability-related structural information, as reflected in the following two aspects. First, they tend to fail to infer the semantics of the code statements with complex logic such as those containing multiple operators and pointers. Second, they are hard to comprehend various code execution sequences, which is essential for precise vulnerability detection. To mitigate the challenges, we propose a Structured Natural Language Comment tree-based vulnerAbiLity dEtection framework based on the pre-trained models, named . The proposed Structured Natural Language Comment Tree (SCT) integrates the semantics of code statements with code execution sequences based on the Abstract Syntax Trees (ASTs).Specifically, comprises three main modules: (1) Comment Tree Construction, which aims at enhancing the model’s ability to infer the semantics of code statements by first incorporating Large Language Models (LLMs) for comment generation and then adding the comment node to ASTs. (2) Structured Natural Language Comment Tree Construction, which aims at explicitly involving code execution sequence by combining the code syntax templates with the comment tree. (3) SCT-Enhanced Representation, which finally incorporates the constructed SCTs for well capturing vulnerability patterns. Experimental results demonstrate that outperforms the best-performing baseline, including the pre-trained model and LLMs, with improvements of 2.96%, 13.47%, and 3.75% in terms of F1 score on the FFMPeg+Qemu, Reveal, and SVulD datasets, respectively. Furthermore, can be applied to different pre-trained models, such as CodeBERT and UniXcoder, yielding the F1 score performance enhancements ranging from 1.37% to 10.87%.

Microservice resilience, the ability of microservices to recover from failures and continue providing reliable and responsive services, is crucial for cloud vendors. However, the current practice relies on manually configured rules specific to a certain microservice system, resulting in labor-intensity and flexibility issues, given the large scale and high dynamics of microservices. A more labor-efficient and versatile solution is desired. Our insight is that resilient deployment can effectively prevent the dissemination of degradation from system performance metrics to user-aware metrics, and the latter directly affects service quality. In other words, failures in a non-resilient deployment can impact both types of metrics, leading to user dissatisfaction. With this in mind, we propose MicroRes, the first versatile resilience profiling framework for microservices via degradation dissemination indexing. MicroRes first injects failures into microservices and collects available monitoring metrics. Then, it ranks the metrics according to their contributions to the overall service degradation. It produces a resilience index by how much the degradation is disseminated from system performance metrics to user-aware metrics. Higher degradation dissemination indicates lower resilience. We evaluate MicroRes on two open-source and one industrial microservice system. The experiments show MicroRes' efficient and effective resilience profiling of microservices. We also showcase MicroRes' practical usage in production.

Data wrangling, the process of preparing raw data for further analysis in computational notebooks, is a crucial yet time-consuming step in data science. Code generation has the potential to automate the data wrangling process to reduce analysts' overhead by translating user intents into executable code. Precisely generating data wrangling code necessitates a comprehensive consideration of the rich context present in notebooks, including textual context, code context and data context. However, notebooks often interleave multiple non-linear analysis tasks into linear sequence of code blocks, where the contextual dependencies are not clearly reflected. Directly training models with source code blocks fails to fully exploit the contexts for accurate wrangling code generation.

API suggestion is a critical task in modern software development, assisting programmers by predicting and recommending third-party APIs based on the current context. Recent advancements in large code models (LCMs) have shown promise in the API suggestion task. However, they mainly focus on suggesting which APIs to use, ignoring that programmers may demand more assistance while using APIs in practice including when to use the suggested APIs and how to use the APIs. To mitigate the gap, we conduct a systematic evaluation of LCMs for the API suggestion task in the paper.

Smart homes, powered by the Internet of Things, offer great convenience but also pose security concerns due to abnormal behaviors, such as improper operations of users and potential attacks from malicious attackers. Several behavior modeling methods have been proposed to identify abnormal behaviors and mitigate potential risks. However, their performance often falls short because they do not effectively learn less frequent behaviors, consider temporal context, or account for the impact of noise in human behaviors. In this paper, we propose SmartGuard, an autoencoder-based unsupervised user behavior anomaly detection framework. First, we design a Loss-guided Dynamic Mask Strategy (LDMS) to encourage the model to learn less frequent behaviors, which are often overlooked during learning. Second, we propose a Three-level Time-aware Position Embedding (TTPE) to incorporate temporal information into positional embedding to detect temporal context anomaly. Third, we propose a Noise-aware Weighted Reconstruction Loss (NWRL) that assigns different weights for routine behaviors and noise behaviors to mitigate the interference of noise behaviors during inference. Comprehensive experiments on three datasets with ten types of anomaly behaviors demonstrates that SmartGuard consistently outperforms state-of-the-art baselines and also offers highly interpretable results.

Image generation models can generate or edit images from a given text. Recent advancements in image generation technology, exemplified by DALL-E and Midjourney, have been groundbreaking. These advanced models, despite their impressive capabilities, are often trained on massive Internet datasets, making them susceptible to generating content that perpetuates social stereotypes and biases, which can lead to severe consequences. Prior research on assessing bias within image generation models suffers from several shortcomings, including limited accuracy, reliance on extensive human labor, and lack of comprehensive analysis. In this paper, we propose BiasPainter, a novel evaluation framework that can accurately, automatically and comprehensively trigger social bias in image generation models. BiasPainter uses a diverse range of seed images of individuals and prompts the image generation models to edit these images using gender, race, and age-neutral queries. These queries span 62 professions, 39 activities, 57 types of objects, and 70 personality traits. The framework then compares the edited images to the original seed images, focusing on the significant changes related to gender, race, and age. BiasPainter adopts a key insight that these characteristics should not be modified when subjected to neutral prompts. Built upon this design, BiasPainter can trigger the social bias and evaluate the fairness of image generation models. We use BiasPainter to evaluate six widely-used image generation models, such as stable diffusion and Midjourney. Experimental results show that BiasPainter can successfully trigger social bias in image generation models. According to our human evaluation, BiasPainter can achieve 90.8% accuracy on automatic bias detection, which is significantly higher than the results reported in previous work.

Evaluating Large Language Models’ (LLMs) anthropomorphic capabilities has become increasingly important in contemporary discourse. Utilizing the emotion appraisal theory from psychology, we propose to evaluate the empathy ability of LLMs, i.e., how their feelings change when presented with specific situations. After a careful and comprehensive survey, we collect a dataset containing over 400 situations that have proven effective in eliciting the eight emotions central to our study. Categorizing the situations into 36 factors, we conduct a human evaluation involving more than 1,200 subjects worldwide. With the human evaluation results as references, our evaluation includes seven LLMs, covering both commercial and open-source models, including variations in model sizes, featuring the latest iterations, such as GPT-4, Mixtral-8x22B, and LLaMA-3.1. We find that, despite several misalignments, LLMs can generally respond appropriately to certain situations. Nevertheless, they fall short in alignment with the emotional behaviors of human beings and cannot establish connections between similar situations. Our collected dataset of situations, the human evaluation results, and the code of our testing framework, i.e., EmotionBench, are publicly available at https://github.com/CUHK-ARISE/EmotionBench.

Disaggregated memory (DM) is a widely discussed datacenter architecture in academia and industry. It decouples computing and memory resources from monolithic servers into two network-connected resource pools. Range indexes are widely adopted by storage systems on DM to efficiently locate and query remote data. However, existing range indexes on DM suffer from either high computing-side cache consumption or high memory-side read amplifications. In this paper, we propose CHIME, a hybrid index combining B+ trees with hopscotch hashing, to achieve low cache consumption and low read amplifications simultaneously. There are three challenges in constructing CHIME on DM, i.e., the complicated optimistic synchronization, the extra metadata access, and the read amplifications introduced by hopscotch hashing. CHIME leverages 1) a three-level optimistic synchronization scheme to synchronize read and write operations with various granularities, 2) an access-aggregated metadata management technique to eliminate extra metadata accesses by piggybacking and replicating metadata, and 3) an effective hotness-aware speculative read mechanism to mitigate the read amplifications of hopscotch hashing. Experimental results show that CHIME outperforms the state-of-the-art range indexes on DM by up to 5.1× with the same cache size and achieves similar performance with up to 8.7× lower cache consumption.

Zero-shottranslation is a promising direction for building a comprehensive multilingual neural machine translation (MNMT) system. However, its quality is still not satisfactory due to off-target issues. In this paper, we aim to understand and alleviate the off-target issues from the perspective of uncertainty in zero-shot translation. By carefully examining the translation output and model confidence, we identify two uncertainties that are responsible for the off-target issues, namely, extrinsic data uncertainty and intrinsic model uncertainty. Based on the observations, we propose two lightweight and complementary approaches to denoise the training data for model training and explicitly penalize the off-target translations by unlikelihood training during model training. Extensive experiments on both balanced and imbalanced datasets show that our approaches significantly improve the performance of zero-shot translation over strong MNMT baselines.

Disaggregated memory (DM) is an increasingly prevalent architecture with high resource utilization. It separates computing and memory resources into two pools and interconnects them with fast networks. Existing range indexes on DM are based on B+ trees, which suffer from large inherent read and write amplifications. The read and write amplifications rapidly saturate the network bandwidth, resulting in low request throughput and high access latency of B+ trees on DM. In this article, we propose that the radix tree is more suitable for DM than the B+ tree due to smaller read and write amplifications. However, constructing a radix tree on DM is challenging due to the costly lock-based concurrency control, the bounded memory-side IOPS, and the complicated computing-side cache validation. To address these challenges, we design SMART , the first radix tree for disaggregated memory with high performance. Specifically, we leverage (1) a hybrid concurrency control scheme including lock-free internal nodes and fine-grained lock-based leaf nodes to reduce lock overhead, (2) a computing-side read-delegation and write-combining technique to break through the IOPS upper bound by reducing redundant I/Os, and (3) a simple yet effective reverse check mechanism for computing-side cache validation. Experimental results show that SMART achieves 6.1× higher throughput under typical write-intensive workloads and 2.8× higher throughput under read-only workloads in YCSB benchmarks, compared with state-of-the-art B+ trees on DM.

Software-based distributed shared memory (DSM) allows multiple processes to access shared data without the need for specialized hardware. However, this flexibility comes at a significant cost due to the need for data synchronization. One approach to mitigate these costs is to relax the consistency model, which can lead to delayed updates to the shared data. This approach typically requires the use of explicit synchronization primitives to regulate access to the shared memory and determine the timing of data synchronization. To circumvent the need for explicit synchronization, an alternative approach is to manage shared memory transparently using the underlying system. While this can simplify programming, it often imposes a fixed granularity for data sharing, which can limit the expansion of the coherence domain and increase the synchronization requirements. To overcome this limitation, we propose an abstraction called the elastic coherence domain, which dynamically adjusts the scope of data synchronization and is supported by the underlying system for transparent management of shared memory. The experimental results show that this approach can improve the efficiency of memory sharing in distributed environments.

Automated logging statement generation supports developers in documenting critical software runtime behavior. While substantial recent research has focused on retrieval-based and learning-based methods, results suggest they fail to provide appropriate logging statements in real-world complex software. Given the great success in natural language generation and programming language comprehension, large language models (LLMs) might help developers generate logging statements, but this has not yet been investigated. To fill the gap, this paper performs the first study on exploring LLMs for logging statement generation. We first build a logging statement generation dataset,LogBench, with two parts: (1)LogBench-O:3,870methods with6,849logging statements collected from GitHub repositories, and (2)LogBench-T: the transformed unseen code from LogBench-O. Then, we leverage LogBench to evaluate theeffectivenessandgeneralization capabilities(usingLogBench-T) of 13 top-performing LLMs, from 60M to 405B parameters. In addition, we examine the performance of these LLMs against classical retrieval-based and machine learning-based logging methods from the era preceding LLMs. Specifically, we evaluate the logging effectiveness of LLMs by studying their ability to determine logging ingredients and the impact of prompts and external program information. We further evaluate LLM's logging generalization capabilities using unseen data (LogBench-T) derived from code transformation techniques. While existing LLMs deliver decent predictions on logging levels and logging variables, our study indicates that they only achieve a maximum BLEU score of0.249, thus calling for improvements. The paper also highlights the importance of prompt constructions and external factors (e.g., programming contexts and code comments) for LLMs’ logging performance. In addition, we observed that existing LLMs show a significant performance drop (8.2%-16.2%decrease) when dealing with logging unseen code, revealing their unsatisfactory generalization capabilities. Based on these findings, we identify five implications and provide practical advice for future logging research. Our empirical analysis discloses the limitations of current logging approaches while showcasing the potential of LLM-based logging tools, and provides actionable guidance for building more practical models.

Advances in Artificial Intelligence (AI), and in particular in Machine Learning (ML), are introducing profound changes to scholarly submissions across publication venues, affecting in particular the contributions that are being submitted to Software Engineering (SE) conferences and journals. In this context, it is not always clear whether manuscripts submitted to SE venues under the umbrella term SE for AI are indeed relevant to SE, in the sense that they explicitly contain contributions to the SE body of knowledge. This leads to recurring discussions on whether certain AI-related submissions are appropriate to SE venues, or should instead be submitted to other journals and conferences, including AI or ML-specific ones. In this editorial, we discuss the kinds of AI-related contributions that are a better fit-and a less good fit-for publication in the IEEE Transactions on Software Engineering.

Despite the excellent performance, deep neural networks (DNNs) have been shown to be vulnerable to adversarial examples. Besides, these examples are often transferable among different models. In other words, the same adversarial example can fool multiple models with different architectures at the same time. Based on this property, many black-box transfer-based attack techniques have been developed. However, current transfer-based attacks generally focus on the cross-architecture setting, where the attacker has access to the training data of the target model, which is not guaranteed in realistic situations. In this paper, we design a Cross-Domain Transfer-Based Attack (CDTA), which works in the cross-domain scenario. In this setting, attackers have no information about the target model, such as its architecture and training data. Specifically, we propose a contrastive spectral training method to train a feature extractor on a source domain (e.g., ImageNet) and use it to craft adversarial examples on target domains (e.g., Oxford 102 Flower). Our method corrupts the semantic information of the benign image by scrambling the outputs of both the intermediate feature layers and the final layer of the feature extractor. We evaluate CDTA with 16 target deep models on four datasets with widely varying styles. The results confirm that, in terms of the attack success rate, our approach can consistently outperform the state-of-the-art baselines by an average of 11.45% across all target models. Our code is available at https://github.com/LiulietLee/CDTA.

The vast amount of sensitive data in modern web applications has become a prime target for cyberattacks. Existing browser security policies disallow the execution of unknown scripts, but do not restrict access to sensitive web content by 'trusted' third-party scripts. Therefore, the over-privileged third-party scripts can compromise the confidentiality and integrity of sensitive user data in the applications.

Vision transformers (ViTs) have been successfully deployed in a variety of computer vision tasks, but they are still vulnerable to adversarial samples. Transfer-based attacks use a local model to generate adversarial samples and directly transfer them to attack a target black-box model. The high efficiency of transfer-based attacks makes it a severe security threat to ViT-based applications. Therefore, it is vital to design effective transfer-based attacks to identify the deficiencies of ViTs beforehand in security-sensitive scenarios. Existing efforts generally focus on regularizing the input gradients to stabilize the updated direction of adversarial samples. However, the variance of the back-propagated gradients in intermediate blocks of ViTs may still be large, which may make the generated adversarial samples focus on some model-specific features and get stuck in poor local optima. To overcome the shortcomings of existing approaches, we propose the Token Gradient Regularization (TGR) method. According to the structural characteristics of ViTs, TGR reduces the variance of the back-propagated gradient in each internal block of ViTs in a token-wise manner and utilizes the regularized gradient to generate adversarial samples. Extensive experiments on attacking both ViTs and CNNs confirm the superiority of our approach. Notably, compared to the state-of-the-art transfer-based at-tacks, our TGR offers a performance improvement of 8.8% on average.

Deep neural networks have achieved unprecedented success on diverse vision tasks. However, they are vulnerable to adversarial noise that is imperceptible to humans. This phenomenon negatively affects their deployment in real-world scenarios, especially security-related ones. To evaluate the robustness of a target model in practice, transfer-based attacks craft adversarial samples with a local model and have attracted increasing attention from researchers due to their high efficiency. The state-of-the-art transfer-based attacks are generally based on data augmentation, which typically augments multiple training images from a linear path when learning adversarial samples. However, such methods selected the image augmentation path heuristically and may augment images that are semantics-inconsistent with the target images, which harms the transferability of the generated adversarial samples. To overcome the pitfall, we propose the Path-Augmented Method (PAM). Specifically, PAM first constructs a candidate augmentation path pool. It then settles the employed augmentation paths during adversarial sample generation with greedy search. Furthermore, to avoid augmenting semantics-inconsistent images, we train a Semantics Predictor (SP) to constrain the length of the augmentation path. Extensive experiments confirm that PAM can achieve an improvement of over 4.8% on average compared with the state-of-the-art baselines in terms of the attack success rates.

As the model size of pre-trained language models (PLMs) grows rapidly, full fine-tuning becomes prohibitively expensive for model training and storage. In vision-and-language (VL), parameter-efficient tuning (PET) techniques are proposed to integrate modular modifications (e.g., Adapter and LoRA) into encoder-decoder PLMs. By tuning a small set of trainable parameters, these techniques perform on par with full fine-tuning. However, excessive modular modifications and neglecting the functionality gap between the encoders and decoders can lead to performance degradation, while existing PET techniques (e.g., VL-Adapter) overlook these critical issues. In this paper, we propose a Vision-and-Language Parameter-Efficient Tuning (VL-PET) framework to impose effective control over modular modifications via a novel granularity-controlled mechanism. Considering different granularity-controlled matrices generated by this mechanism, a variety of model-agnostic VL-PET modules can be instantiated from our framework for better efficiency and effective-ness trade-offs. We further propose lightweight PET module designs to enhance VL alignment and modeling for the encoders and maintain text generation for the decoders. Extensive experiments conducted on four image-text tasks and four video-text tasks demonstrate the efficiency, effectiveness and transferability of our VL-PET framework. In particular, our VL-PETlargewith lightweight PET module designs significantly outperforms VL-Adapter by 2.92% (3.41%) and LoRA by 3.37% (7.03%) with BART-base (T5-base) on image-text tasks. Furthermore, we validate the enhanced effect of employing our VL-PET designs on existing PET techniques, enabling them to achieve significant performance improvements. Our code is available at https://github.com/HenryHZY/VL-PET.

Logs, being run-time information automatically generated by software, record system events and activities with their timestamps. Before obtaining more insights into the run-time status of the software, a fundamental step of log analysis, called log parsing, is employed to extract structured templates and parameters from the semi-structured raw log messages. However, current log parsers are all syntax-based and regard each message as a character string, ignoring the semantic information included in parameters and templates. Thus, we propose the first semantic-based parser SemParser to unlock the critical bottleneck of mining semantics from log messages. It contains two steps, an end-to-end semantics miner and a joint parser. Specifically, the first step aims to identify explicit semantics inside a single log, and the second step is responsible for jointly inferring implicit semantics and computing structural outputs according to the contextual knowledge base of the logs. To analyze the effectiveness of our semantic parser, we first demonstrate that it can derive rich semantics from log messages collected from six widely-applied systems with an average F1 score of 0.985. Then, we conduct two representative downstream tasks, showing that current downstream models improve their performance with appropriately extracted semantics by 1.2%-11.7% and 8.65% on two anomaly detection datasets and a failure identification dataset, respectively. We believe these findings provide insights into semantically understanding log messages for the log analysis community.

The complexity and dynamism of microservices pose significant challenges to system reliability, and thereby, automated troubleshooting is crucial. Effective root cause localization after anomaly detection is crucial for ensuring the reliability of microservice systems. However, two significant issues rest in existing approaches: (1) Microservices generate traces, system logs, and key performance indicators (KPIs), but existing approaches usually consider traces only, failing to understand the system fully as traces cannot depict all anomalies; (2) Troubleshooting microservices generally contains two main phases, i.e., anomaly detection and root cause localization. Existing studies regard these two phases as independent, ignoring their close correlation. Even worse, inaccurate detection results can deeply affect localization effectiveness. To overcome these limitations, we propose Eadro, the first end-to-end framework to integrate anomaly detection and root cause localization based on multi-source data for troubleshooting large-scale microservices. The key insights of Eadro are the anomaly manifestations on different data sources and the close connection between detection and localization. Thus, Eadro models intra-service behaviors and inter-service dependencies from traces, logs, and KPIs, all the while leveraging the shared knowledge of the two phases via multi-task learning. Experiments on two widely-used benchmark microservices demonstrate that Eadro outperforms state-of-the-art approaches by a large margin. The results also show the usefulness of integrating multi-source data. We also release our code and data to facilitate future research.

Prompt and accurate detection of system anomalies is essential to ensure the reliability of software systems. Unlike manual efforts that exploit all available run-time information, existing approaches usually leverage only a single type of monitoring data (often logs or metrics) or fail to make effective use of the joint information among different types of data. Consequently, many false predictions occur. To better understand the manifestations of system anomalies, we conduct a systematical study on a large amount of heterogeneous data, i.e., logs and metrics. Our study demonstrates that logs and metrics can manifest system anomalies collaboratively and complementarily, and neither of them only is sufficient. Thus, integrating heterogeneous data can help recover the complete picture of a system's health status. In this context, we propose Hades, the first end-to-end semi-supervised approach to effectively identify system anomalies based on heterogeneous data. Our approach employs a hierarchical architecture to learn a global representation of the system status by fusing log semantics and metric patterns. It captures discriminative features and meaningful interactions from heterogeneous data via a cross-modal attention module, trained in a semi-supervised manner. We evaluate Hades extensively on large-scale simulated data and datasets from Huawei Cloud. The experimental results present the effectiveness of our model in detecting system anomalies. We also release the code and the annotated dataset for replication and future research.

In cloud systems, incidents are potential threats to customer satisfaction and business revenue. When customers are affected by incidents, they often request customer support service (CSS) from the cloud provider by submitting a support ticket. Many tickets could be duplicate as they are reported in a distributed and uncoordinated manner. Thus, aggregating such duplicate tickets is essential for efficient ticket management. Previous studies mainly rely on tickets' textual similarity to detect duplication; however, duplicate tickets in a cloud system could carry semantically different descriptions due to the complex service dependency of the cloud system. To tackle this problem, we propose iPACK, an incident-aware method for aggregating duplicate tickets by fusing the failure information between the customer side (i.e., tickets) and the cloud side (i.e., incidents). We extensively evaluate iPACK on three datasets collected from the production environment of a large-scale cloud platform, Azure. The experimental results show that iPACK can precisely and comprehensively aggregate duplicate tickets, achieving an F1 score of 0.871~0.935 and outperforming state-of-the-art methods by 12.4%~31.2%.

The exponential growth of social media platforms such as Twitter and Facebook has revolutionized textual communication and textual content publication in human society. However, they have been increasingly exploited to propagate toxic content, such as hate speech, malicious advertisement, and pornography, which can lead to highly negative impacts (e.g., harmful effects on teen mental health). Researchers and practitioners have been enthusiastically developing and extensively deploying textual content moderation software to address this problem. However, we find that malicious users can evade moderation by changing only a few words in the toxic content. Moreover, modern content moderation software's performance against malicious inputs remains underexplored. To this end, we propose MTTM, a Metamorphic Testing framework for Textual content Moderation software. Specifically, we conduct a pilot study on 2, 000 text messages collected from real users and summarize eleven metamorphic relations across three perturbation levels: character, word, and sentence. MTTM employs these metamorphic relations on toxic textual contents to generate test cases, which are still toxic yet likely to evade moderation. In our evaluation, we employ MTTM to test three commercial textual content moderation software and two state-of-the-art moderation algorithms against three kinds of toxic content. The results show that MTTM achieves up to 83.9%, 51%, and 82.5% error finding rates (EFR) when testing commercial moderation software provided by Google, Baidu, and Huawei, respectively, and it obtains up to 91.2% EFR when testing the state-of-the-art algorithms from the academy. In addition, we leverage the test cases generated by MTTM to retrain the model we explored, which largely improves model robustness 0% ~ 5.9% EFR) while maintaining the accuracy on the original test set. A demo can be found in this link1.

Despite their stunning performance, developing deep learning models from scratch is a formidable task. Therefore, it popularizes Machine-Learning-as-a-Service (MLaaS), where general users can access the trained models of MLaaS providers via Application Programming Interfaces (APIs) on a pay-per-query basis. Unfortunately, the success of MLaaS is under threat from model extraction attacks, where attackers intend to extract a local model of equivalent functionality to the target MLaaS model. However, existing studies on model extraction of text analytics APIs frequently assume adversaries have strong knowledge about the victim model, like its architecture and parameters, which hardly holds in practice. Besides, since the attacker's and the victim's training data can be considerably discrepant, it is non-trivial to perform efficient model extraction. In this paper, to advance the understanding of such attacks, we propose a framework, PEEP, for practical and efficient model extraction of sentiment analysis APIs with only query access. Specifically, PEEP features a learning-based scheme, which employs out-of-domain public corpora and a novel query strategy to construct proxy training data for model extraction. Besides, PEEP introduces a greedy search algorithm to settle an appropriate architecture for the extracted model. We conducted extensive experiments with two victim models across three datasets and two real-life commercial sentiment analysis APIs. Experimental results corroborate that PEEP can consistently outperform the state-of-the-art baselines in terms of effectiveness and efficiency.

Language models are known to be vulnerable to textual adversarial attacks, which add human-imperceptible perturbations to the input to mislead DNNs. It is thus imperative to devise effective attack algorithms to identify the deficiencies of DNNs before real-world deployment. However, existing word-level attacks have two major deficiencies: (1) They may change the semantics of the original sentence. (2) The generated adversarial sample can appear unnatural to humans due to the introduction of out-of-domain substitute words. In this paper, to address such drawbacks, we propose a semantics- and domain-aware word-level attack method. Specifically, we greedily replace the important words in a sentence with the ones suggested by a language model. The language model is trained to be semantics- and domain-aware via contrastive learning and in-domain pre-training. Furthermore, to balance the quality of adversarial examples and the attack success rate, we propose an iterative updating framework to optimize the contrastive learning loss and the in-domain pre-training loss in circular order. Comprehensive experimental comparisons confirm the superiority of our approach. Notably, compared with state-of-the-art benchmarks, our strategy can achieve over 3\% improvement in attack success rates and 9.8\% improvement in the quality of adversarial examples.

Software logs record system activities, aiding maintainers in identifying the underlying causes for failures and enabling prompt mitigation actions. However, maintainers need to inspect a large volume of daily logs to identify the anomalous logs that reveal failure details for further diagnosis. Thus, how to automatically distinguish these anomalous logs from normal logs becomes a critical problem. Existing approaches alleviate the burden on software maintainers, but they are built upon an improper yet critical assumption: logging statements in the software remain unchanged. While software keeps evolving, our empirical study finds that evolving software brings three challenges: log parsing errors, evolving log events, and unstable log sequences. In this paper, we propose a novel unsupervised approach named Evolving Log analyzer (EvLog) to mitigate these challenges. We first build a multi-level representation extractor to process logs without parsing to prevent errors from the parser. The multi-level representations preserve the essential semantics of logs while leaving out insignificant changes in evolving events. EvLog then implements an anomaly discriminator with an attention mechanism to identify the anomalous logs and avoid the issue brought by the unstable sequence. EvLog has shown effectiveness in two real-world system evolution log datasets with an average F1 score of 0.955 and 0.847 in the intra-version setting and inter-version setting, respectively, which outperforms other state-of-the-art approaches by a wide margin. To our best knowledge, this is the first study on localizing anomalous logs over software evolution. We believe our work sheds new light on the impact of software evolution with the corresponding solutions for the log analysis community.

As modern software systems continue to grow in terms of complexity and volume, anomaly detection on multivariate monitoring metrics, which profile systems’ health status, becomes more and more critical and challenging. In particular, the dependency between different metrics and their historical patterns plays a critical role in pursuing prompt and accurate anomaly detection. Existing approaches fall short of industrial needs for being unable to capture such information efficiently. To fill this significant gap, in this paper, we propose CMAnomaly, an anomaly detection framework on multivariate monitoring metrics based on collaborative machine. The proposed collaborative machine is a mechanism to capture the pairwise interactions along with feature and temporal dimensions with linear time complexity. Cost-effective models can then be employed to leverage both the dependency between monitoring metrics and their historical patterns for anomaly detection. The proposed framework is extensively evaluated with both public data and industrial data collected from a large-scale online service system of Huawei Cloud. The experimental results demonstrate that compared with state-of-the-art baseline models, CMAnomaly achieves an average F1 score of 0.9494, outperforming baselines by 6.77% ~ 10.68%, and runs 10× ~ 20× faster. Furthermore, we also share our experience of deploying CMAnomaly in Huawei Cloud.

Logs have been widely adopted in software system development and maintenance because of the rich runtime information they record. In recent years, the increase of software size and complexity leads to the rapid growth of the volume of logs. To handle these large volumes of logs efficiently and effectively, a line of research focuses on developing intelligent and automated log analysis techniques. However, only a few of these techniques have reached successful deployments in industry due to the lack of public log datasets and open benchmarking upon them. To fill this significant gap and facilitate more research on AI-driven log analytics, we have collected and released loghub, a large collection of system log datasets. In particular, loghub provides 19 real-world log datasets collected from a wide range of software systems, including distributed systems, supercomputers, operating systems, mobile systems, server applications, and standalone software. In this paper, we summarize the statistics of these datasets, introduce some practical usage scenarios of the loghub datasets, and present our benchmarking results on loghub to benefit the researchers and practitioners in this field. Up to the time of this paper writing, the loghub datasets have been downloaded for roughly 90,000 times in total by hundreds of organizations from both industry and academia. The loghub datasets are available at https://github.com/logpai/loghub.

The exponential growth of social media platforms, such as Facebook, Instagram, Youtube, and TikTok, has revolutionized communication and content publication in human society. Users on these platforms can publish multimedia content that delivers information via the combination of text, audio, images, and video. Meanwhile, the multimedia content release facility has been increasingly exploited to propagate toxic content, such as hate speech, malicious advertisement, and pornography. To this end, content moderation software has been widely deployed on these platforms to detect and blocks toxic content. However, due to the complexity of content moderation models and the difficulty of understanding information across multiple modalities, existing content moderation software can fail to detect toxic content, which often leads to extremely negative impacts (e.g., harmful effects on teen mental health). We introduce Semantic Fusion, a general, effective methodology for validating multimedia content moderation software. Our key idea is to fuse two or more existing single-modal inputs (e.g., a textual sentence and an image) into a new input that combines the semantics of its ancestors in a novel manner and has toxic nature by construction. This fused input is then used for validating multimedia content moderation software. We realized Semantic Fusion as DUO, a practical content moderation software testing tool. In our evaluation, we employ DUO to test five commercial content moderation software and two state-of-the-art models against three kinds of toxic contents. The results show that DUO achieves up to 100% error finding rate (EFR) when testing moderation software and it obtains up to 94.1% EFR when testing the state-of-the-art models. In addition, we leverage the test cases generated by DUO to retrain the two models we explored, which largely improves model robustness (2.5%∼5.7% EFR) while maintaining the accuracy on the original test set.

Pre-trained models of source code have gained widespread popularity in many code intelligence tasks. Recently, with the scaling of the model and corpus size, large language models have shown the ability of in-context learning (ICL). ICL employs task instructions and a few examples as demonstrations, and then inputs the demonstrations to the language models for making predictions. This new learning paradigm is training-free and has shown impressive performance in various natural language processing and code intelligence tasks. However, the performance of ICL heavily relies on the quality of demonstrations, e.g., the selected examples. It is important to systematically investigate how to construct a good demonstration for code-related tasks. In this paper, we empirically explore the impact of three key factors on the performance of ICL in code intelligence tasks: the selection, order, and number of demonstration examples. We conduct extensive experiments on three code intelligence tasks including code summarization, bug fixing, and program synthesis. Our experimental results demonstrate that all the above three factors dramatically impact the performance of ICL in code intelligence tasks. Additionally, we summarize our findings and provide takeaway suggestions on how to construct effective demonstrations, taking into account these three perspectives. We also show that a carefully-designed demonstration based on our findings can lead to substantial improvements over widely-used demonstration construction methods, e.g., improving BLEU-4, EM, and EM by at least 9.90%, 175.96%, and 50.81% on code summarization, bug fixing, and program synthesis, respectively.

The rapid progress of modern computing systems has led to a growing interest in informative run-time logs. Various log-based anomaly detection techniques have been proposed to ensure software reliability. However, their implementation in the industry has been limited due to the lack of high-quality public log resources as training datasets. While some log datasets are available for anomaly detection, they suffer from limitations in (1) comprehensiveness of log events; (2) scalability over diverse systems; and (3) flexibility of log utility. To address these limitations, we propose AUTOLOG, the first automated log generation methodology for anomaly detection. AUTOLOG uses program analysis to generate runtime log sequences without actually running the system. AUTOLOG starts with probing comprehensive logging statements associated with the call graphs of an application. Then, it constructs execution graphs for each method after pruning the call graphs to find log-related execution paths in a scalable manner. Finally, AUTOLOG propagates the anomaly label to each acquired execution path based on human knowledge. It generates flexible log sequences by walking along the log execution paths with controllable parameters. Experiments on 50 popular Java projects show that AUTOLOG acquires significantly more (9x-58x) log events than existing log datasets from the same system, and generates log messages much faster (15x) with a single machine than existing passive data collection approaches. AUTOLOG also provides hyper-parameters to adjust the data size, anomaly rate, and component indicator for simulating different real-world scenarios. We further demonstrate AUTOLOG's practicality by showing that AUTOLOG enables log-based anomaly detectors to achieve better performance (1.93%) compared to existing log datasets. We hope AUTOLOG can facilitate the benchmarking and adoption of automated log analysis techniques.

WebAssembly (Wasm) is a bytecode format originally serving as a compilation target for Web applications. It has recently been used increasingly on the server side, e.g., providing a safer, faster, and more portable alternative to Linux containers. With the popularity of server-side Wasm applications, it is essential to study performance issues (i.e., abnormal latency) in Wasm runtimes, as they may cause a significant impact on server-side applications. However, there is still a lack of attention to performance issues in server-side Wasm runtimes. In this paper, we design a novel differential testing approach WarpDiff to identify performance issues in server-side Wasm runtimes. The key insight is that in normal cases, the execution time of the same test case on different Wasm runtimes should follow an oracle ratio. We identify abnormal cases where the execution time ratio significantly deviates from the oracle ratio and subsequently locate the Wasm runtimes that cause the performance issues. We apply WarpDiff to test five popular server-side Wasm runtimes using 123 test cases from the LLVM test suite and demonstrate the top 10 abnormal cases we identified. We further conduct an in-depth analysis of these abnormal cases and summarize seven performance issues, all of which have been confirmed by the developers. We hope our work can inspire future investigation on improving Wasm runtime implementation and thus promoting the development of server-side Wasm applications.

Ensuring the reliability and user satisfaction of cloud services necessitates prompt anomaly detection followed by diagnosis. Existing techniques for anomaly detection focus solely on real-time detection, meaning that anomaly alerts are issued as soon as anomalies occur. However, anomalies can propagate and escalate into failures, making faster-than-real-time anomaly detection highly desirable for expediting downstream analysis and intervention. This paper proposes Maat, the first work to address anomaly anticipation of performance metrics in cloud services. Maat adopts a novel two-stage paradigm for anomaly anticipation, consisting of metric forecasting and anomaly detection on forecasts. The metric forecasting stage employs a conditional denoising diffusion model to enable multi-step forecasting in an auto-regressive manner. The detection stage extracts anomaly-indicating features based on domain knowledge and applies isolation forest with incremental learning to detect upcoming anomalies. Thus, our method can uncover anomalies that better conform to human expertise. Evaluation on three publicly available datasets demonstrates that Maat can anticipate anomalies faster than real-time comparatively or more effectively compared with state-of-the-art real-time anomaly detectors. We also present cases highlighting Maat's success in forecasting abnormal metrics and discovering anomalies.

Ensuring the reliability of cloud systems is critical for both cloud vendors and customers. Cloud systems often rely on virtualization techniques to create instances of hardware resources, such as virtual machines. However, virtualization hinders the observability of cloud systems, making it challenging to diagnose platform-level issues. To improve system observability, we propose to infer functional clusters of instances, i.e., groups of instances having similar functionalities. We first conduct a pilot study on a large-scale cloud system, i.e., Huawei Cloud, demonstrating that instances having similar functionalities share similar communication and resource usage patterns. Motivated by these findings, we formulate the identification of functional clusters as a clustering problem and propose a non-intrusive solution called Prism. Prism adopts a coarse-to-fine clustering strategy. It first partitions instances into coarse-grained chunks based on communication patterns. Within each chunk, Prism further groups instances with similar resource usage patterns to produce fine-grained functional clusters. Such a design reduces noises in the data and allows Prism to process massive instances efficiently. We evaluate Prism on two datasets collected from the real-world production environment of Huawei Cloud. Our experiments show that Prism achieves a v-measure of ∼0.95, surpassing existing state-of-the-art solutions. Additionally, we illustrate the integration of Prism within monitoring systems for enhanced cloud reliability through two real-world use cases.

Python is a popular dynamic programming language, evidenced by its ranking as the second most commonly used language on GitHub. However, its dynamic type system can lead to potential type errors, leading researchers to explore automatic type inference approaches for Python programs. Existing type inference approaches can be generally grouped into three categories, i.e., rule-based, supervised, and cloze- style approaches. The rule-based type inference approaches can ensure the accuracy of predicted variable types, but they suffer from low coverage problems caused by dynamic features and external calls. Supervised type inference approaches, while feature-agnostic and able to mitigate the low coverage problem, require large, high- quality annotated datasets and are limited to pre-defined types. As zero-shot approaches, the cloze-style approaches reformulate the type inference problem into a fill-in-the-blank problem by leveraging the general knowledge in powerful pre-trained code models. However, their performance is limited since they ignore the domain knowledge from static typing rules which reflect the inference logic. What is more, their predictions are not interpretable, hindering developers' understanding and verification of the results. This paper introduces Typegen, a few-shot generative type inference approach that incorporates static domain knowledge from static analysis. Typegen creates chain-of-thought (COT) prompts by translating the type inference steps of static analysis into prompts based on the type dependency graphs (TDGs), enabling language models to learn from how static analysis infers types. By combining COT prompts with code slices and type hints, TypegEnconstructs example prompts from human annotations. Typeg Enonly requires very few annotated examples to teach language models to generate similar COT prompts via in-context learning. Moreover, Typeg Enenhances the interpretability of results through the use of the input- explanation-output strategy, which generates both explanations and type predictions in COT prompts. Experiments show that Typegen outperforms the best baseline Type4Py by 10.0% for argument type prediction and 22.5 % in return value type prediction in terms of top-l Exact Match by using only five examples. Furthermore, Typeg Enachieves substantial improvements of 27 % to 84 % compared to the zero-shot performance of large language models with parameter sizes ranging from 1.3B to 175B in terms of top-I Exact Match.

The exponential growth of social media platforms has brought about a revolution in communication and content dissemination in human society. Nevertheless, these platforms are being increasingly misused to spread toxic content, including hate speech, malicious advertising, and pornography, leading to severe negative consequences such as harm to teenagers' mental health. Despite tremendous efforts in developing and deploying textual and image content moderation methods, malicious users can evade moderation by embedding texts into images, such as screenshots of the text, usually with some interference. We find that modern content moderation software's performance against such malicious inputs remains underexplored. In this work, we propose OASIS, a metamorphic testing framework for content moderation software. OASIS employs 21 transform rules summarized from our pilot study on 5,000 real-world toxic contents collected from 4 popular social media applications, including Twitter, Instagram, Sina Weibo, and Baidu Tieba. Given toxic textual contents, OASIS can generate image test cases, which preserve the toxicity yet are likely to bypass moderation. In the evaluation, we employ OASIS to test five commercial textual content moderation software from famous companies (i.e., Google Cloud, Microsoft Azure, Baidu Cloud, Alibaba Cloud and Tencent Cloud), as well as a state-of-the-art moderation research model. The results show that OASIS achieves up to 100% error finding rates. Moreover, through retraining the models with the test cases generated by OASIS, the robustness of the moderation model can be improved without performance degradation.

Software plays a crucial role in our daily lives, and therefore the quality and security of software systems have become increasingly important. However, vulnerabilities in software still pose a significant threat, as they can have serious consequences. Recent advances in automated program repair have sought to automatically detect and fix bugs using data-driven techniques. Sophisticated deep learning methods have been applied to this area and have achieved promising results. However, existing benchmarks for training and evaluating these techniques remain limited, as they tend to focus on a single programming language and have relatively small datasets. Moreover, many benchmarks tend to be outdated and lack diversity, focusing on a specific codebase. Worse still, the quality of bug explanations in existing datasets is low, as they typically use imprecise and uninformative commit messages as explanations. To address these issues, we propose an automated collecting framework REEF to collect REal-world vulnErabilities and Fixes from open-source repositories. We focus on vulnerabilities since they are exploitable and have serious consequences. We develop a multi-language crawler to collect vulnerabilities and their fixes, and design metrics to filter for high-quality vulnerability-fix pairs. Furthermore, we propose a neural language model-based approach to generate high-quality vulnerability explanations, which is key to producing informative fix messages. Through extensive experiments, we demonstrate that our approach can collect high-quality vulnerability-fix pairs and generate strong explanations. The dataset we collect contains 4,466 CVEs with 30,987 patches (including 236 CWE) across 7 programming languages with detailed related information, which is superior to existing benchmarks in scale, coverage, and quality. Evaluations by human experts further confirm that our framework produces high-quality vulnerability explanations.

Search-based software testing (SBST) has proved its effectiveness in generating test cases to achieve its defined test goals, such as branch and data-dependency coverage. However, to detect more program faults in an effective way, pre-defined goals can hardly be adaptive in diversified projects. In this work, we propose KAT, a novel knowledge-aware SBST approach to generate on-demand assertions in the program under test (PUT) based on its used APIs. KAT constructs an API knowledge graph from the API documentation to derive the constraints that the client codes need to satisfy. Each constraint is instrumented into the PUT as a program branch, serving as a test goal to guide SBST to detect faults. We evaluate KAT with two baselines (i.e., EvoSuite and Catcher) with a close-world and an open-world experiment to detect API bugs. The close-world experiment shows that KAT outperforms the baselines in the F1-score (0.55 vs. 0.24 and 0.30) to detect API-related bugs. The open-world experiment shows that KAT can detect 59.64% and 9.05% more bugs than the baselines in practice.

Powered by advanced Artificial Intelligence (AI) techniques, conversational AI systems, such as ChatGPT, and digital assistants like Siri, have been widely deployed in daily life. However, such systems may still produce content containing biases and stereotypes, causing potential social problems. Due to modern AI techniques’ data-driven, black-box nature, comprehensively identifying and measuring biases in conversational systems remains challenging. Particularly, it is hard to generate inputs that can comprehensively trigger potential bias due to the lack of data containing both social groups and biased properties. In addition, modern conversational systems can produce diverse responses (e.g., chatting and explanation), which makes existing bias detection methods based solely on sentiment and toxicity hardly being adopted. In this paper, we propose BiasAsker, an automated framework to identify and measure social bias in conversational AI systems. To obtain social groups and biased properties, we construct a comprehensive social bias dataset containing a total of 841 groups and 5,021 biased properties. Given the dataset, BiasAsker automatically generates questions and adopts a novel method based on existence measurement to identify two types of biases (i.e., absolute bias and related bias) in conversational systems. Extensive experiments on eight commercial systems and two famous research models, such as ChatGPT and GPT-3, show that 32.83% of the questions generated by BiasAsker can trigger biased behaviors in these widely deployed conversational systems. All the code, data, and experimental results have been released to facilitate future research.

In-memory caching systems are fundamental building blocks in cloud services. However, due to the coupled CPU and memory on monolithic servers, existing caching systems cannot elastically adjust resources in a resource-efficient and agile manner. To achieve better elasticity, we propose to port in-memory caching systems to the disaggregated memory (DM) architecture, where compute and memory resources are decoupled and can be allocated flexibly. However, constructing an elastic caching system on DM is challenging since accessing cached objects with CPU-bypass remote memory accesses hinders the execution of caching algorithms. Moreover, the elastic changes of compute and memory resources on DM affect the access patterns of cached data, compromising the hit rates of caching algorithms. We design Ditto, the first caching system on DM, to address these challenges. Ditto first proposes a client-centric caching framework to efficiently execute various caching algorithms in the compute pool of DM, relying only on remote memory accesses. Then, Ditto employs a distributed adaptive caching scheme that adaptively switches to the best-fit caching algorithm in real-time based on the performance of multiple caching algorithms to improve cache hit rates. Our experiments show that Ditto effectively adapts to the changing resources on DM and outperforms the state-of-the-art caching systems by up to 3.6× in real-world workloads and 9× in YCSB benchmarks.

Abstract Mobile apps are becoming an integral part of people's daily life by providing various functionalities, such as messaging and gaming. App developers try their best to ensure user experience during app development and maintenance to improve the rating of their apps on app platforms and attract more user downloads. Previous studies indicated that responding to users' reviews tends to change their attitude towards the apps positively. Users who have been replied are more likely to update the given ratings. However, reading and responding to every user review is not an easy task for developers since it is common for popular apps to receive tons of reviews every day. Thus, automation tools for review replying are needed. To address the need above, the paper introduces a Transformer‐based approach, named TRRGen, to automatically generate responses to given user reviews. TRRGen extracts apps' categories, rating, and review text as the input features. By adapting a Transformer‐based model, TRRGen can generate appropriate replies for new reviews. Comprehensive experiments and analysis on the real‐world datasets indicate that the proposed approach can generate high‐quality replies for users' reviews and significantly outperform current state‐of‐art approaches on the task. The manual validation results on the generated replies further demonstrate the effectiveness of the proposed approach.

Code summaries help developers comprehend programs and reduce their time to infer the program functionalities during software maintenance. Recent efforts resort to deep learning techniques such as sequence-to-sequence models for generating accurate code summaries, among which Transformer-based approaches have achieved promising performance. However, effectively integrating the code structure information into the Transformer is under-explored in this task domain. In this article, we propose a novel approach named SG-Trans to incorporate code structural properties into Transformer. Specifically, we inject the local symbolic information (e.g., code tokens and statements) and global syntactic structure (e.g., dataflow graph) into the self-attention module of Transformer as inductive bias. To further capture the hierarchical characteristics of code, the local information and global structure are designed to distribute in the attention heads of lower layers and high layers of Transformer. Extensive evaluation shows the superior performance of SG-Trans over the state-of-the-art approaches. Compared with the best-performing baseline, SG-Trans still improves 1.4% and 2.0% on two benchmark datasets, respectively, in terms of METEOR score, a metric widely used for measuring generation quality.

Developers often need to decide which APIs to use for the functions being implemented. With the ever-growing number of APIs and libraries, it becomes increasingly difficult for developers to find appropriate APIs, indicating the necessity of automatic API usage recommendation. Previous studies adopt statistical models or collaborative filtering methods to mine the implicit API usage patterns for recommendation. However, they rely on the occurrence frequencies of APIs for mining usage patterns, thus prone to fail for the low-frequency APIs. Besides, prior studies generally regard the API call interaction graph as homogeneous graph, ignoring the rich information (e.g., edge types) in the structure graph. In this work, we propose a novel method namedMEGAfor improving the recommendation accuracy especially for the low-frequency APIs. Specifically, besidescall interaction graph, MEGA considers another two new heterogeneous graphs:global API co-occurrence graphenriched with the API frequency information andhierarchical structure graphenriched with the project component information. With the three multi-view heterogeneous graphs, MEGA can capture the API usage patterns more accurately. Experiments on three Java benchmark datasets demonstrate that MEGA significantly outperforms the baseline models by at least 19% with respect to the Success Rate@1 metric. Especially, for the low-frequency APIs, MEGA also increases the baselines by at least 55% regarding the Success Rate@1 score.

Application Programming Interfaces (APIs), which encapsulate the implementation of specific functions as interfaces, greatly improve the efficiency of modern software development. As the number of APIs grows up fast nowadays, developers can hardly be familiar with all the APIs and usually need to search for appropriate APIs for usage. So lots of efforts have been devoted to improving the API recommendation task. However, it has been increasingly difficult to gauge the performance of new models due to the lack of a uniform definition of the task and a standardized benchmark. For example, some studies regard the task as a code completion problem, while others recommend relative APIs given natural language queries. To reduce the challenges and better facilitate future research, in this paper, we revisit the API recommendation task and aim at benchmarking the approaches. Specifically, the paper groups the approaches into two categories according to the task definition, i.e., query-based API recommendation and code-based API recommendation. We study 11 recently-proposed approaches along with 4 widely-used IDEs. One benchmark named APIBench is then built for the two respective categories of approaches. Based on APIBench, we distill some actionable insights and challenges for API recommendation. We also achieve some implications and directions for improving the performance of recommending APIs, including appropriate query reformulation, data source selection, low resource setting, user-defined APIs, and query-based API recommendation with usage patterns.

Pre-trained models have been shown effective in many code intelligence tasks, such as automatic code summarization and defect prediction. These models are pre-trained on large-scale unlabeled corpus and then fine-tuned in downstream tasks. However, as the inputs to pre-training and downstream tasks are in different forms, it is hard to fully explore the knowledge of pre-trained models. Besides, the performance of fine-tuning strongly relies on the amount of downstream task data, while in practice, the data scarcity scenarios are common. Recent studies in the natural language processing (NLP) field show that prompt tuning, a new paradigm for tuning, alleviates the above issues and achieves promising results in various NLP tasks. In prompt tuning, the prompts inserted during tuning provide task-specific knowledge, which is especially beneficial for tasks with relatively scarce data. In this article, we empirically evaluate the usage and effect of prompt tuning in code intelligence tasks. We conduct prompt tuning on popular pre-trained models CodeBERT and CodeT5 and experiment with four code intelligence tasks including defect prediction, code search, code summarization, and code translation. Our experimental results show that prompt tuning consistently outperforms fine-tuning in all four tasks. In addition, prompt tuning shows great potential in low-resource scenarios, e.g., improving the BLEU scores of fine-tuning by more than 26% on average for code summarization. Our results suggest that instead of fine-tuning, we could adapt prompt tuning for code intelligence tasks to achieve better performance, especially when lacking task-specific data. We also discuss the implications for adapting prompt tuning in code intelligence tasks.

Front-running attacks have been a major concern on the blockchain. Attackers launch front-running attacks by inserting additional transactions before upcoming victim transactions to manipulate victim transaction executions and make profits. Recent studies have shown that front-running attacks are prevalent on the Ethereum blockchain and have caused millions of US dollars loss. It is the vulnerabilities in smart contracts, which are blockchain programs invoked by transactions, that enable the front-running attack opportunities. Although techniques to detect front-running vulnerabilities have been proposed, their performance on real-world vulnerable contracts is unclear. There is no large-scale benchmark based on real attacks to evaluate their capabilities. We make four contributions in this paper. First, we design an effective algorithm to mine real-world attacks in the blockchain history. The evaluation shows that our mining algorithm is more effective and comprehensive, achieving higher recall in finding real attacks than the previous study. Second, we propose an automated and scalable vulnerability localization approach to localize code snippets in smart contracts that enable front-running attacks. The evaluation also shows that our localization approaches are effective in achieving higher precision in pinpointing vulnerabilities compared to the baseline technique. Third, we build a benchmark consisting of 513 real-world attacks with vulnerable code labeled in 235 distinct smart contracts, which is useful to help understand the nature of front-running attacks, vulnerabilities in smart contracts, and evaluate vulnerability detection techniques. Last but not least, we conduct an empirical evaluation of seven state-of-the-art vulnerability detection techniques on our benchmark. The evaluation experiment reveals the inadequacy of existing techniques in detecting front-running vulnerabilities, with a low recall of$\leq$6.04%. Our further analysis identifies four common limitations in existing techniques: lack of support for inter-contract analysis, inefficient constraint solving for cryptographic operations, improper vulnerability patterns, and lack of token support.

Text revision refers to a family of natural language generation tasks, where the source and target sequences share moderate resemblance in surface form but differentiate in attributes, such as text formality and simplicity. Current state-of-the-art methods formulate these tasks as sequence-to-sequence learning problems, which rely on large-scale parallel training corpus. In this paper, we present an iterative in-place editing approach for text revision, which requires no parallel data. In this approach, we simply fine-tune a pre-trained Transformer with masked language modeling and attribute classification. During inference, the editing at each iteration is realized by two-step span replacement. At the first step, the distributed representation of the text optimizes on the fly towards an attribute function. At the second step, a text span is masked and another new one is proposed conditioned on the optimized representation. The empirical experiments on two typical and important text revision tasks, text formalization and text simplification, show the effectiveness of our approach. It achieves competitive and even better performance than state-of-the-art supervised methods on text simplification, and gains better performance than strong unsupervised methods on text formalization. Our code and model are released at https://github.com/jingjingli01/OREO.

Wenchao Gu, Yanlin Wang, Lun Du, Hongyu Zhang, Shi Han, Dongmei Zhang, Michael Lyu. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Wenxuan Wang, Wenxiang Jiao, Yongchang Hao, Xing Wang, Shuming Shi, Zhaopeng Tu, Michael Lyu. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Deep neural networks (DNNs) are known to be vulnerable to adversarial examples. It is thus imperative to devise effective attack algorithms to identify the deficiencies of DNNs beforehand in security-sensitive applications. To efficiently tackle the black-box setting where the target model's particulars are unknown, feature-level transfer-based attacks propose to contaminate the intermediate feature outputs of local models, and then directly employ the crafted adversarial samples to attack the target model. Due to the transferability of features, feature-level attacks have shown promise in synthesizing more transferable adversarial samples. However, existing feature-level attacks generally employ inaccurate neuron importance estimations, which deteriorates their transferability. To overcome such pitfalls, in this paper, we propose the Neuron Attribution-based Attack (NAA), which conducts feature-level attacks with more accurate neuron importance estimations. Specifically, we first completely attribute a model's output to each neuron in a middle layer. We then derive an approximation scheme of neuron attribution to tremendously reduce the computation overhead. Finally, we weight neurons based on their attribution results and launch feature-level attacks. Extensive experiments confirm the superiority of our approach to the state-of-the-art benchmarks. Our code is available at: hups.//rgithub.com/jprhang1810/NAA.

Alerts are crucial for requesting prompt human intervention upon cloud anomalies. The quality of alerts significantly affects the cloud reliability and the cloud provider’s business revenue. In practice, we observe on-call engineers being hindered from quickly locating and fixing faulty cloud services because of the vast existence of misleading, non-informative, non-actionable alerts. We call the ineffectiveness of alerts "anti-patterns of alerts". To better understand the anti-patterns of alerts and provide actionable measures to mitigate anti-patterns, in this paper, we conduct the first empirical study on the practices of mitigating anti-patterns of alerts in an industrial cloud system. We study the alert strategies and the alert processing procedure at Huawei Cloud, a leading cloud provider. Our study combines the quantitative analysis of millions of alerts in two years and a survey with eighteen experienced engineers. As a result, we summarized four individual anti-patterns and two collective anti-patterns of alerts. We also summarize four current reactions to mitigate the anti-patterns of alerts, and the general preventative guidelines for the configuration of alert strategy. Lastly, we propose to explore the automatic evaluation of the Quality of Alerts (QoA), including the indicativeness, precision, and handleability of alerts, as a future research direction that assists in the automatic detection of alerts’ anti-patterns. The findings of our study are valuable for optimizing cloud monitoring systems and improving the reliability of cloud services.

Recent advances in large-scale pre-training provide large models with the potential to learn knowledge from the raw text.It is thus natural to ask whether it is possible to leverage these large models as knowledge bases for downstream tasks.In this work, we answer the aforementioned question in unsupervised knowledge-grounded conversation.We explore various methods that best elicit knowledge from large models.Our human study indicates that, though hallucinations exist, large models post the unique advantage of being able to output common sense and summarize facts that cannot be directly retrieved from the search engine.To better exploit such generated knowledge in dialogue generation, we treat the generated knowledge as a noisy knowledge source and propose the posterior-based reweighing as well as the noisy training strategy.Empirical results on two benchmarks show advantages over the state-of-the-art methods.

Despite recent progress in open-domain dialogue evaluation, how to develop automatic metrics remains an open problem.We explore the potential of dialogue evaluation featuring dialog act information, which was hardly explicitly modeled in previous methods.However, defined at the utterance level in general, dialog act is of coarse granularity, as an utterance can contain multiple segments possessing different functions.Hence, we propose segment act, an extension of dialog act from utterance level to segment level, and crowdsource a largescale dataset for it.To utilize segment act flows, sequences of segment acts, for evaluation, we develop the first consensus-based dialogue evaluation framework, FlowEval.This framework provides a reference-free approach for dialog evaluation by finding pseudo-references.Extensive experiments against strong baselines on three benchmark datasets demonstrate the effectiveness and other desirable characteristics of our FlowEval, pointing out a potential path for better dialogue evaluation.

To ensure the performance of online service systems, their status is closely monitored with various software and system metrics. Performance anomalies represent the performance degradation issues (e.g., slow response) of the service systems. When performing anomaly detection over the metrics, existing methods often lack the merit of interpretability, which is vital for engineers and analysts to take remediation actions. Moreover, they are unable to effectively accommodate the ever-changing services in an online fashion. To address these limitations, in this paper, we propose ADSketch, an interpretable and adaptive performance anomaly detection approach based on pattern sketching. ADSketch achieves interpretability by identifying groups of anomalous metric patterns, which represent particular types of performance issues. The underlying issues can then be immediately recognized if similar patterns emerge again. In addition, an adaptive learning algorithm is designed to embrace unprecedented patterns induced by service updates or user behavior changes. The proposed approach is evaluated with public data as well as industrial data collected from a representative online service system in Huawei Cloud. The experimental results show that ADSketch outperforms state-of-the-art approaches by a significant margin, and demonstrate the effectiveness of the online algorithm in new pattern discovery. Furthermore, our approach has been successfully deployed in industrial practice.

Online technical forums (e.g., StackOverflow) are popular platforms for developers to discuss technical problems such as how to use a specific Application Programming Interface (API), how to solve the programming tasks, or how to fix bugs in their code. These discussions can often provide auxiliary knowledge of how to use the software that is not covered by the official documents. The automatic extraction of such knowledge may support a set of downstream tasks like API searching or indexing. However, unlike official documentation written by experts, discussions in open forums are made by regular developers who write in short and informal texts, including spelling errors or abbreviations. There are three major challenges for the accurate APIs recognition and linking mentioned APIs from unstructured natural language documents to an entry in the API repository: (1) distinguishing API mentions from common words; (2) identifying API mentions without a fully qualified name; and (3) disambiguating API mentions with similar method names but in a different library. In this paper, to tackle these challenges, we propose an ARCLIN tool, which can effectively distinguish and link APIs without using human annotations. Specifically, we first design an API recognizer to automatically extract API mentions from natural language sentences by a Conditional Random Field (CRF) on the top of a Bi-directional Long Short-Term Memory (Bi-LSTM) module, then we apply a context-aware scoring mechanism to compute the mention-entry similarity for each entry in an API repository. Compared to previous approaches with heuristic rules, our proposed tool without manual inspection outperforms by 8% in a high-quality dataset Py-mention, which contains 558 mentions and 2,830 sentences from five popular Python libraries. To our best knowledge, ARCLIN is the first approach to achieve full automation of API mention resolution from unformatted text without manually collected labels.

Type inference for dynamic programming languages such as Python is an important yet challenging task. Static type inference techniques can precisely infer variables with enough static constraints but are unable to handle variables with dynamic features. Deep learning (DL) based approaches are feature-agnostic, but they cannot guarantee the correctness of the predicted types. Their performance significantly depends on the quality of the training data (i.e., DL models perform poorly on some common types that rarely appear in the training dataset). It is interesting to note that the static and DL-based approaches offer complementary benefits. Unfortunately, to our knowledge, precise type inference based on both static inference and neural predictions has not been exploited and remains an open challenge. In particular, it is hard to integrate DL models into the framework of rule-based static approaches.

Due to the labor-intensive nature of manual test oracle construction, various automated testing techniques have been proposed to enhance the reliability of Natural Language Processing (NLP) software. In theory, these techniques mutate an existing test case (e.g., a sentence with its label) and assume the generated one preserves an equivalent or similar semantic meaning and thus, the same label. However, in practice, many of the generated test cases fail to preserve similar semantic meaning and are unnatural (e.g., grammar errors), which leads to a high false alarm rate and unnatural test cases. Our evaluation study finds that 44% of the test cases generated by the state-of-the-art (SOTA) approaches are false alarms. These test cases require extensive manual checking effort, and instead of improving NLP software, they can even degrade NLP software when utilized in model training. To address this problem, we propose AEON for Automatic Evaluation Of NLP test cases. For each generated test case, it outputs scores based on semantic similarity and language naturalness. We employ AEON to evaluate test cases generated by four popular testing techniques on five datasets across three typical NLP tasks. The results show that AEON aligns the best with human judgment. In particular, AEON achieves the best average precision in detecting semantic inconsistent test cases, outperforming the best baseline metric by 10%. In addition, AEON also has the highest average precision of finding unnatural test cases, surpassing the baselines by more than 15%. Moreover, model training with test cases prioritized by AEON leads to models that are more accurate and robust, demonstrating AEON’s potential in improving NLP software.

Network quantization has gained increasing attention with the rapid growth of large pre-trained language models~(PLMs). However, most existing quantization methods for PLMs follow quantization-aware training~(QAT) that requires end-to-end training with full access to the entire dataset. Therefore, they suffer from slow training, large memory overhead, and data accessibility issues. In this paper, we study post-training quantization~(PTQ) of PLMs, and propose module-wise quantization error minimization~(MREM), an efficient solution to mitigate these issues. By partitioning the PLM into multiple modules, we minimize the reconstruction error incurred by quantization for each module. In addition, we design a new model parallel training strategy such that each module can be trained locally on separate computing devices without waiting for preceding modules, which brings nearly the theoretical training speed-up (e.g., $4\times$ on $4$ GPUs). Experiments on GLUE and SQuAD benchmarks show that our proposed PTQ solution not only performs close to QAT, but also enjoys significant reductions in training time, memory overhead, and data consumption.

Pre-trained models have been shown effective in many code intelligence tasks. These models are pre-trained on large-scale unlabeled corpus and then fine-tuned in downstream tasks. However, as the inputs to pre-training and downstream tasks are in different forms, it is hard to fully explore the knowledge of pre-trained models. Besides, the performance of fine-tuning strongly relies on the amount of downstream data, while in practice, the scenarios with scarce data are common. Recent studies in the natural language processing (NLP) field show that prompt tuning, a new paradigm for tuning, alleviates the above issues and achieves promising results in various NLP tasks. In prompt tuning, the prompts inserted during tuning provide task-specific knowledge, which is especially beneficial for tasks with relatively scarce data. In this paper, we empirically evaluate the usage and effect of prompt tuning in code intelligence tasks. We conduct prompt tuning on popular pre-trained models CodeBERT and CodeT5 and experiment with three code intelligence tasks including defect prediction, code summarization, and code translation. Our experimental results show that prompt tuning consistently outperforms fine-tuning in all three tasks. In addition, prompt tuning shows great potential in low-resource scenarios, e.g., improving the BLEU scores of fine-tuning by more than 26% on average for code summarization. Our results suggest that instead of fine-tuning, we could adapt prompt tuning for code intelligence tasks to achieve better performance, especially when lacking task-specific data.

We present DistillFlow, a knowledge distillation approach to learning optical flow. DistillFlow trains multiple teacher models and a student model, where challenging transformations are applied to the input of the student model to generate hallucinated occlusions as well as less confident predictions. Then, a self-supervised learning framework is constructed: confident predictions from teacher models are served as annotations to guide the student model to learn optical flow for those less confident predictions. The self-supervised learning framework enables us to effectively learn optical flow from unlabeled data, not only for non-occluded pixels, but also for occluded pixels. DistillFlow achieves state-of-the-art unsupervised learning performance on both KITTI and Sintel datasets. Our self-supervised pre-trained model also provides an excellent initialization for supervised fine-tuning, suggesting an alternate training paradigm in contrast to current supervised learning methods that highly rely on pre-training on synthetic data. At the time of writing, our fine-tuned models ranked 1st among all monocular methods on the KITTI 2015 benchmark, and outperform all published methods on the Sintel Final benchmark. More importantly, we demonstrate the generalization capability of DistillFlow in three aspects: framework generalization, correspondence generalization and cross-dataset generalization. Our code and models will be available on https://github.com/ppliuboy/DistillFlow.

Recent years have witnessed the success of natural language generation (NLG) accomplished by deep neural networks, which require a large amount of training data for optimization. With the constant increase of data scale, the complex patterns and potential noises make training NLG models difficult. In order to fully utilize large-scale training data, we explore inactive examples in the training data and propose to rejuvenate the inactive examples for improving the performance of NLG models. Specifically, we define inactive examples as those sentence pairs that contribute less to the performance of NLG models, and show that their existence is independent of model variants but mainly determined by the data distribution. We further introducedata rejuvenationto improve the training of NLG models by re-labeling the inactive examples. The rejuvenated examples and active examples are combined to train a final NLG model. We evaluate our approach by experiments on machine translation (MT) and text summarization (TS) tasks, and achieve significant improvements of performance. Extensive analyses reveal that inactive examples are more difficult to learn than active ones and rejuvenation can reduce the learning difficulty, which stabilizes and accelerates the training process of NLG models and results in models with better generalization capability.

User experience of mobile apps is an essential ingredient that can influence the user base and app revenue. To ensure good user experience and assist app development, several prior studies resort to analysis of app reviews, a type of repository that directly reflects user opinions about the apps. Accurately responding to the app reviews is one of the ways to relieve user concerns and thus improve user experience. However, the response quality of the existing method relies on the pre-extracted features from other tools, including manually labelled keywords and predicted review sentiment, which may hinder the generalizability and flexibility of the method. In this article, we propose a novel neural network approach, named CoRe, with the contextual knowledge naturally incorporated and without involving external tools. Specifically, CoRe integrates two types of contextual knowledge in the training corpus, including official app descriptions from app store and responses of the retrieved semantically similar reviews, for enhancing the relevance and accuracy of the generated review responses. Experiments on practical review data show that CoRe can outperform the state-of-the-art method by 12.36% in terms of BLEU-4, an accuracy metric that is widely used to evaluate text generation systems.

Rust is an emerging programming language that aims at preventing memory-safety bugs without sacrificing much efficiency. The claimed property is very attractive to developers, and many projects start using the language. However, can Rust achieve the memory-safety promise? This article studies the question by surveying 186 real-world bug reports collected from several origins, which contain all existing Rust common vulnerability and exposures (CVEs) of memory-safety issues by 2020-12-31. We manually analyze each bug and extract their culprit patterns. Our analysis result shows that Rust can keep its promise that all memory-safety bugs require unsafe code, and many memory-safety bugs in our dataset are mild soundness issues that only leave a possibility to write memory-safety bugs without unsafe code. Furthermore, we summarize three typical categories of memory-safety bugs, including automatic memory reclaim, unsound function, and unsound generic or trait. While automatic memory claim bugs are related to the side effect of Rust newly-adopted ownership-based resource management scheme, unsound function reveals the essential challenge of Rust development for avoiding unsound code, and unsound generic or trait intensifies the risk of introducing unsoundness. Based on these findings, we propose two promising directions toward improving the security of Rust development, including several best practices of using specific APIs and methods to detect particular bugs involving unsafe code. Our work intends to raise more discussions regarding the memory-safety issues of Rust and facilitate the maturity of the language.

With the growing number of competing Web services that provide similar functionality, Quality-of-Service (QoS) prediction is becoming increasingly important for various QoS-aware approaches of Web services. Collaborative filtering (CF), which is among the most successful personalized prediction techniques for recommender systems, has been widely applied to Web service QoS prediction. In addition to using conventional CF techniques, a number of studies extend the CF approach by incorporating additional information about services and users, such as location, time, and other contextual information from the service invocations. There are also some studies that address other challenges in QoS prediction, such as adaptability, credibility, privacy preservation, and so on. In this survey, we summarize and analyze the state-of-the-art CF QoS prediction approaches of Web services and discuss their features and differences. We also present several Web service QoS datasets that have been used as benchmarks for evaluating the predition accuracy and outline some possible future research directions.

Millions of mobile apps are available in app stores, such as Apple's App Store and Google Play. For a mobile app, it would be increasingly challenging to stand out from the enormous competitors and become prevalent among users. Good user experience and well-designed functionalities are the keys to a successful app. To achieve this, popular apps usually schedule their updates frequently. If we can capture the critical app issues faced by users in a timely and accurate manner, developers can make timely updates, and good user experience can be ensured. There exist prior studies on analyzing reviews for detecting emerging app issues. These studies are usually based on topic modeling or clustering techniques. However, the short-length characteristics and sentiment of user reviews have not been considered. In this paper, we propose a novel emerging issue detection approach named MERIT to take into consideration the two aforementioned characteristics. Specifically, we propose an Adaptive Online Biterm Sentiment-Topic (AOBST) model for jointly modeling topics and corresponding sentiments that takes into consideration app versions. Based on the AOBST model, we infer the topics negatively reflected in user reviews for one app version, and automatically interpret the meaning of the topics with most relevant phrases and sentences. Experiments on popular apps from Google Play and Apple's App Store demonstrate the effectiveness of MERIT in identifying emerging app issues, improving the state-of-the-art method by 22.3 percent in terms of F1-score. In terms of efficiency, MERIT can return results within acceptable time.

Haoli Bai, Wei Zhang, Lu Hou, Lifeng Shang, Jin Jin, Xin Jiang, Qun Liu, Michael Lyu, Irwin King. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Wenxiang Jiao, Xing Wang, Zhaopeng Tu, Shuming Shi, Michael Lyu, Irwin King. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Although deep neural networks (DNNs) have achieved tremendous performance in diverse vision challenges, they are surprisingly susceptible to adversarial examples, which are born of intentionally perturbing benign samples in a human-imperceptible fashion. It thus poses security concerns on the deployment of DNNs in practice, particularly in safety- and security-sensitive domains. To investigate the robustness of DNNs, transfer-based attacks have attracted a growing interest recently due to their high practical applicability, where attackers craft adversarial samples with local models and employ the resultant samples to attack a remote black-box model. However, existing transfer-based attacks frequently suffer from low success rates due to overfitting to the adopted local model. To boost the transferability of adversarial samples, we propose to improve the robustness of synthesized adversarial samples via adversarial transformations. Specifically, we employ an adversarial transformation network to model the most harmful distortions that can destroy adversarial noises and require the synthesized adversarial samples to become resistant to such adversarial transformations. Extensive experiments on the ImageNet benchmark showcase the superiority of our method to state-of-the-art baselines in attacking both undefended and defended models.

Function-as-a-Service (FaaS) is becoming a prevalent paradigm in developing cloud applications. With FaaS, clients can develop applications as serverless functions, leaving the burden of resource management to cloud providers. However, FaaS platforms suffer from the performance degradation caused by the cold starts of serverless functions. Cold starts happen when serverless functions are invoked before they have been loaded into the memory. The problem is unavoidable because the memory in datacenters is typically too limited to hold all serverless functions simultaneously. The latency of cold function invocations will greatly degenerate the performance of FaaS platforms. Currently, FaaS platforms employ various scheduling methods to reduce the occurrences of cold starts. However, they do not consider the ubiquitous dependencies between serverless functions. Observing the potential of using dependencies to mitigate cold starts, we propose Defuse, a Dependency-guided Function Scheduler on FaaS platforms. Specifically, Defuse identifies two types of dependencies between serverless functions, i.e., strong dependencies and weak ones. It uses frequent pattern mining and positive point-wise mutual information to mine such dependencies respectively from function invocation histories. In this way, Defuse constructs a function dependency graph. The connected components (i.e., dependent functions) on the graph can be scheduled to diminish the occurrences of cold starts. We evaluate the effectiveness of Defuse by applying it to an industrial serverless dataset. The experimental results show that Defuse can reduce 22% of memory usage while having a 35% decrease in function cold-start rates compared with the state-of-the-art method.

Variational autoencoders (VAEs) have been widely applied for text modeling. In practice, however, they are troubled by two challenges: information underrepresentation and posterior collapse. The former arises as only the last hidden state of LSTM encoder is transformed into the latent space, which is generally insufficient to summarize the data. The latter is a longstanding problem during the training of VAEs as the optimization is trapped to a disastrous local optimum. In this paper, we propose Discrete Auto-regressive Variational Attention Model (DAVAM) to address the challenges. Specifically, we introduce an auto-regressive variational attention approach to enrich the latent space by effectively capturing the semantic dependency from the input. We further design discrete latent space for the variational attention and mathematically show that our model is free from posterior collapse. Extensive experiments on language modeling tasks demonstrate the superiority of DAVAM against several VAE counterparts. Code will be released.

As online service systems continue to grow in terms of complexity and volume, how service incidents are managed will significantly impact company revenue and user trust. Due to the cascading effect, cloud failures often come with an overwhelming number of incidents from dependent services and devices. To pursue efficient incident management, related incidents should be quickly aggregated to narrow down the problem scope. To this end, in this paper, we propose GRLIA, an incident aggregation framework based on graph representation learning over the cascading graph of cloud failures. A representation vector is learned for each unique type of incident in an unsupervised and unified manner, which is able to simultaneously encode the topological and temporal correlations among incidents. Thus, it can be easily employed for online incident aggregation. In particular, to learn the correlations more accurately, we try to recover the complete scope of failures’ cascading impact by leveraging fine-grained system monitoring data, i.e., Key Performance Indicators (KPIs). The proposed framework is evaluated with real-world incident data collected from a large-scale online service system of Huawei Cloud. The experimental results demonstrate that GRLIA is effective and outperforms existing methods. Furthermore, our framework has been successfully deployed in industrial practice.

Service reliability is one of the key challenges that cloud providers have to deal with. In cloud systems, unplanned service failures may cause severe cascading impacts on their dependent services, deteriorating customer satisfaction. Predicting the cascading impacts accurately and efficiently is critical to the operation and maintenance of cloud systems. Existing approaches identify whether one service depends on another via distributed tracing but no prior work focused on discriminating to what extent the dependency between cloud services is. In this paper, we survey the outages and the procedure for failure diagnosis in two cloud providers to motivate the definition of the intensity of dependency. We define the intensity of dependency between two services as how much the status of the callee service influences the caller service. Then we propose AID, the first approach to predict the intensity of dependencies between cloud services. AID first generates a set of candidate dependency pairs from the spans. AID then represents the status of each cloud service with a multivariate time series aggregated from the spans. With the representation of services, AID calculates the similarities between the statuses of the caller and the callee of each candidate pair. Finally, AID aggregates the similarities to produce a unified value as the intensity of the dependency. We evaluate AID on the data collected from an open-source microservice benchmark and a cloud system in production. The experimental results show that AID can efficiently and accurately predict the intensity of dependencies. We further demonstrate the usefulness of our method in a large-scale commercial cloud system.

Yongchang Hao, Shilin He, Wenxiang Jiao, Zhaopeng Tu, Michael Lyu, Xing Wang. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Data-dependent hashing methods have demonstrated good performance in various machine learning applications to learn a low-dimensional representation from the original data. However, they still suffer from several obstacles: First, most of existing hashing methods are trained in a batch mode, yielding inefficiency for training streaming data. Second, the computational cost and the memory consumption increase extraordinarily in the big data setting, which perplexes the training procedure. Third, the lack of labeled data hinders the improvement of the model performance. To address these difficulties, we utilize online sketching hashing (OSH) and present a FasteR Online Sketching Hashing (FROSH) algorithm to sketch the data in a more compact form via an independent transformation. We provide theoretical justification to guarantee that our proposed FROSH consumes less time and achieves a comparable sketching precision under the same memory cost of OSH. We also extend FROSH to its distributed implementation, namely DFROSH, to further reduce the training time cost of FROSH while deriving the theoretical bound of the sketching precision. Finally, we conduct extensive experiments on both synthetic and real datasets to demonstrate the attractive merits of FROSH and DFROSH.

Machine learning–(ML) based approach is considered as one of the most promising techniques for Android malware detection and has achieved high accuracy by leveraging commonly used features. In practice, most of the ML classifications only provide a binary label to mobile users and app security analysts. However, stakeholders are more interested in the reason why apps are classified as malicious in both academia and industry. This belongs to the research area of interpretable ML but in a specific research domain (i.e., mobile malware detection). Although several interpretable ML methods have been exhibited to explain the final classification results in many cutting-edge Artificial Intelligent–based research fields, until now, there is no study interpreting why an app is classified as malware or unveiling the domain-specific challenges. In this article, to fill this gap, we propose a novel and interpretable ML-based approach (named XMal ) to classify malware with high accuracy and explain the classification result meanwhile. (1) The first classification phase of XMal hinges multi-layer perceptron and attention mechanism and also pinpoints the key features most related to the classification result. (2) The second interpreting phase aims at automatically producing neural language descriptions to interpret the core malicious behaviors within apps. We evaluate the behavior description results by leveraging a human study and an in-depth quantitative analysis. Moreover, we further compare XMal with the existing interpretable ML-based methods (i.e., Drebin and LIME) to demonstrate the effectiveness of XMal . We find that XMal is able to reveal the malicious behaviors more accurately. Additionally, our experiments show that XMal can also interpret the reason why some samples are misclassified by ML classifiers. Our study peeks into the interpretable ML through the research of Android malware detection and analysis.

Model compression has been widely adopted to obtain light-weighted deep neural networks. Most prevalent methods, however, require fine-tuning with sufficient training data to ensure accuracy, which could be challenged by privacy and security issues. As a compromise between privacy and performance, in this paper we investigate few shot network compression: given few samples per class, how can we effectively compress the network with negligible performance drop? The core challenge of few shot network compression lies in high estimation errors from the original network during inference, since the compressed network can easily over-fits on the few training instances. The estimation errors could propagate and accumulate layer-wisely and finally deteriorate the network output. To address the problem, we propose cross distillation, a novel layer-wise knowledge distillation approach. By interweaving hidden layers of teacher and student network, layer-wisely accumulated estimation errors can be effectively reduced. The proposed method offers a general framework compatible with prevalent network compression techniques such as pruning. Extensive experiments n benchmark datasets demonstrate that cross distillation can significantly improve the student network's accuracy when only a few training instances are available.

Real-time emotion recognition (RTER) in conversations is significant for developing emotionally intelligent chatting machines. Without the future context in RTER, it becomes critical to build the memory bank carefully for capturing historical context and summarize the memories appropriately to retrieve relevant information. We propose an Attention Gated Hierarchical Memory Network (AGHMN) to address the problems of prior work: (1) Commonly used convolutional neural networks (CNNs) for utterance feature extraction are less compatible in the memory modules; (2) Unidirectional gated recurrent units (GRUs) only allow each historical utterance to have context before it, preventing information propagation in the opposite direction; (3) The Soft Attention for summarizing loses the positional and ordering information of memories, regardless of how the memory bank is built. Particularly, we propose a Hierarchical Memory Network (HMN) with a bidirectional GRU (BiGRU) as the utterance reader and a BiGRU fusion layer for the interaction between historical utterances. For memory summarizing, we propose an Attention GRU (AGRU) where we utilize the attention weights to update the internal state of GRU. We further promote the AGRU to a bidirectional variant (BiAGRU) to balance the contextual information from recent memories and that from distant memories. We conduct experiments on two emotion conversation datasets with extensive analysis, demonstrating the efficacy of our AGHMN models.

Recent NLP studies reveal that substantial linguistic information can be attributed to single neurons, i.e., individual dimensions of the representation vectors. We hypothesize that modeling strong interactions among neurons helps to better capture complex information by composing the linguistic properties embedded in individual neurons. Starting from this intuition, we propose a novel approach to compose representations learned by different components in neural machine translation (e.g., multi-layer networks or multi-head attention), based on modeling strong interactions among neurons in the representation vectors. Specifically, we leverage bilinear pooling to model pairwise multiplicative interactions among individual neurons, and a low-rank approximation to make the model computationally feasible. We further propose extended bilinear pooling to incorporate first-order representations. Experiments on WMT14 English⇒German and English⇒French translation tasks show that our model consistently improves performances over the SOTA Transformer baseline. Further analyses demonstrate that our approach indeed captures more syntactic and semantic information as expected.

Yifan Gao, Chien-Sheng Wu, Shafiq Joty, Caiming Xiong, Richard Socher, Irwin King, Michael Lyu, Steven C.H. Hoi. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020.

In this paper, we propose a unified method to jointly learn optical flow and stereo matching. Our first intuition is stereo matching can be modeled as a special case of optical flow, and we can leverage 3D geometry behind stereoscopic videos to guide the learning of these two forms of correspondences. We then enroll this knowledge into the state-of-the-art self-supervised learning framework, and train one single network to estimate both flow and stereo. Second, we unveil the bottlenecks in prior self-supervised learning approaches, and propose to create a new set of challenging proxy tasks to boost performance. These two insights yield a single model that achieves the highest accuracy among all existing unsupervised flow and stereo methods on KITTI 2012 and 2015 benchmarks. More remarkably, our self-supervised method even outperforms several state-of-the-art fully supervised methods, including PWC-Net and FlowNet2 on KITTI 2012.

The widespread deployment of deep models necessitates the assessment of model vulnerability in practice, especially for safety- and security-sensitive domains such as autonomous driving and medical diagnosis. Transfer-based attacks against image classifiers thus elicit mounting interest, where attackers are required to craft adversarial images based on local proxy models without the feedback information from remote target ones. However, under such a challenging but practical setup, the synthesized adversarial samples often achieve limited success due to overfitting to the local model employed. In this work, we propose a novel mechanism to alleviate the overfitting issue. It computes model attention over extracted features to regularize the search of adversarial examples, which prioritizes the corruption of critical features that are likely to be adopted by diverse architectures. Consequently, it can promote the transferability of resultant adversarial instances. Extensive experiments on ImageNet classifiers confirm the effectiveness of our strategy and its superiority to state-of-the-art benchmarks in both white-box and black-box settings.

With the growing prevalence of convolutional neural networks (CNNs), there is an urgent demand to explain their behaviors. Global explanations contribute to understanding model predictions on a whole category of samples, and thus have attracted increasing interest recently. However, existing methods overwhelmingly conduct separate input attribution or rely on local approximations of models, making them fail to offer faithful global explanations of CNNs. To overcome such drawbacks, we propose a novel two-stage framework, Attacking for Interpretability (AfI), which explains model decisions in terms of the importance of user-defined concepts. AfI first conducts a feature occlusion analysis, which resembles a process of attacking models to derive the category-wide importance of different features. We then map the feature importance to concept importance through ad-hoc semantic tasks. Experimental results confirm the effectiveness of AfI and its superiority in providing more accurate estimations of concept importance than existing proposals.

Yifan Gao, Chien-Sheng Wu, Jingjing Li, Shafiq Joty, Steven C.H. Hoi, Caiming Xiong, Irwin King, Michael Lyu. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020.

Large-scale training datasets lie at the core of the recent success of neural machine translation (NMT) models.However, the complex patterns and potential noises in the large-scale data make training NMT models difficult.In this work, we explore to identify the inactive training examples which contribute less to the model performance, and show that the existence of inactive examples depends on the data distribution.We further introduce data rejuvenation to improve the training of NMT models on large-scale datasets by exploiting inactive examples.The proposed framework consists of three phases.First, we train an identification model on the original training data, and use it to distinguish inactive examples and active examples by their sentence-level output probabilities.Then, we train a rejuvenation model on the active examples, which is used to re-label the inactive examples with forwardtranslation.Finally, the rejuvenated examples and the active examples are combined to train the final NMT model.Experimental results on WMT14 English-German and English-French datasets show that the proposed data rejuvenation consistently and significantly improves performance for several strong NMT models.Extensive analyses reveal that our approach stabilizes and accelerates the training process of NMT models, resulting in final models with better generalization capability. 1

Visual dialog is a challenging vision-language task, where a dialog agent needs to answer a series of questions through reasoning on the image content and dialog history.Prior work has mostly focused on various attention mechanisms to model such intricate interactions.By contrast, in this work, we propose VD-BERT, a simple yet effective framework of unified vision-dialog Transformer that leverages the pretrained BERT language models for Visual Dialog tasks.The model is unified in that (1) it captures all the interactions between the image and the multi-turn dialog using a single-stream Transformer encoder, and (2) it supports both answer ranking and answer generation seamlessly through the same architecture.More crucially, we adapt BERT for the effective fusion of vision and dialog contents via visually grounded training.Without the need of pretraining on external vision-language data, our model yields new state of the art, achieving the top position in both single-model and ensemble settings (74.54 and 75.35NDCG scores) on the visual dialog leaderboard.Our code and pretrained models are released at https: //github.com/salesforce/VD-BERT.

Social media produces large amounts of contents every day.To help users quickly capture what they need, keyphrase prediction is receiving a growing attention.Nevertheless, most prior efforts focus on text modeling, largely ignoring the rich features embedded in the matching images.In this work, we explore the joint effects of texts and images in predicting the keyphrases for a multimedia post.To better align social media style texts and images, we propose: (1) a novel Multi-Modality Multi-Head Attention (M 3 H-Att) to capture the intricate cross-media interactions; (2) image wordings, in forms of optical characters and image attributes, to bridge the two modalities.Moreover, we design a novel unified framework to leverage the outputs of keyphrase classification and generation and couple their advantages.Extensive experiments on a large-scale dataset 1 newly collected from Twitter show that our model significantly outperforms the previous state of the art based on traditional co-attentions.Further analyses show that our multi-head attention is able to attend information from various aspects and boost classification or generation in diverse scenarios.

In this work, we propose TGLS, a novel framework for unsupervised Text Generation by Learning from Search. We start by applying a strong search algorithm (in particular, simulated annealing) towards a heuristically defined objective that (roughly) estimates the quality of sentences. Then, a conditional generative model learns from the search results, and meanwhile smooth out the noise of search. The alternation between search and learning can be repeated for performance bootstrapping. We demonstrate the effectiveness of TGLS on two real-world natural language generation tasks, unsupervised paraphrasing and text formalization. Our model significantly outperforms unsupervised baseline methods in both tasks. Especially, it achieves comparable performance to strong supervised methods for paraphrase generation.

Recent advances in neural architecture search inspire many channel number search algorithms~(CNS) for convolutional neural networks. To improve searching efficiency, parameter sharing is widely applied, which reuses parameters among different channel configurations. Nevertheless, it is unclear how parameter sharing affects the searching process. In this paper, we aim at providing a better understanding and exploitation of parameter sharing for CNS. Specifically, we propose affine parameter sharing~(APS) as a general formulation to unify and quantitatively analyze existing channel search algorithms. It is found that with parameter sharing, weight updates of one architecture can simultaneously benefit other candidates. However, it also results in less confidence in choosing good architectures. We thus propose a new strategy of parameter sharing towards a better balance between training efficiency and architecture discrimination. Extensive analysis and experiments demonstrate the superiority of the proposed strategy in channel configuration against many state-of-the-art counterparts on benchmark datasets.

The management of cloud service incidents (unplanned interruptions or outages of a service/product) greatly affects customer satisfaction and business revenue. After years of efforts, cloud enterprises are able to solve most incidents automatically and timely. However, in practice, we still observe critical service incidents that occurred in an unexpected manner and orchestrated diagnosis workflow failed to mitigate them. In order to accelerate the understanding of unprecedented incidents and provide actionable recommendations, modern incident management system employs the strategy of AIOps (Artificial Intelligence for IT Operations). In this paper, to provide a broad view of industrial incident management and understand the modern incident management system, we conduct a comprehensive empirical study spanning over two years of incident management practices at Microsoft. Particularly, we identify two critical challenges (namely, incomplete service/resource dependencies and imprecise resource health assessment) and investigate the underlying reasons from the perspective of cloud system design and operations. We also present IcM BRAIN, our AIOps framework towards intelligent incident management, and show its practical benefits conveyed to the cloud services of Microsoft.

In our world with full of uncertainty, debates and argumentation contribute to the progress of science and society. Despite of the increasing attention to characterize human arguments, most progress made so far focus on the debate outcome, largely ignoring the dynamic patterns in argumentation processes. This paper presents a study that automatically analyzes the key factors in argument persuasiveness, beyond simply predicting who will persuade whom. Specifically, we propose a novel neural model that is able to dynamically track the changes of latent topics and discourse in argumentative conversations, allowing the investigation of their roles in influencing the outcomes of persuasion. Extensive experiments have been conducted on argumentative conversations on both social media and supreme court. The results show that our model outperforms state-of-the-art models in identifying persuasive arguments via explicitly exploring dynamic factors of topic and discourse. We further analyze the effects of topics and discourse on persuasiveness, and find that they are both useful — topics provide concrete evidence while superior discourse styles may bias participants, especially in social media arguments. In addition, we draw some findings from our empirical results, which will help people better engage in future persuasive conversations.

Abstract Software obfuscation has been developed for over 30 years. A problem always confusing the communities is what security strength the technique can achieve. Nowadays, this problem becomes even harder as the software economy becomes more diversified. Inspired by the classic idea of layered security for risk management, we propose layered obfuscation as a promising way to realize reliable software obfuscation. Our concept is based on the fact that real-world software is usually complicated. Merely applying one or several obfuscation approaches in an ad-hoc way cannot achieve good obscurity. Layered obfuscation, on the other hand, aims to mitigate the risks of reverse software engineering by integrating different obfuscation techniques as a whole solution. In the paper, we conduct a systematic review of existing obfuscation techniques based on the idea of layered obfuscation and develop a novel taxonomy of obfuscation techniques. Following our taxonomy hierarchy, the obfuscation strategies under different branches are orthogonal to each other. In this way, it can assist developers in choosing obfuscation techniques and designing layered obfuscation solutions based on their specific requirements.

Symbolic execution has become an indispensable technique for software testing and program analysis. However, since several symbolic execution tools are presently available off-the-shelf, there is a need for a practical benchmarking approach. This paper introduces a fresh approach that can help benchmark symbolic execution tools in a fine-grained and efficient manner. The approach evaluates the performance of such tools against known challenges faced by general symbolic execution techniques, e.g., floating-point numbers and symbolic memories. We first survey related papers and systematize the challenges of symbolic execution. We extract 12 distinct challenges from the literature and categorize them into two categories: symbolic-reasoning challenges and path-explosion challenges. Next, we develop a dataset of logic bombs and a framework for benchmarking symbolic execution tools automatically. For each challenge, our dataset contains several logic bombs, each addressing a specific challenging problem. Triggering one or more logic bombs confirms that the symbolic execution tool in question is able to handle the corresponding problem. Real-world experiments with three popular symbolic execution tools, namely, KLEE, angr, and Triton have shown that our approach can reveal the capabilities and limitations of the tools in handling specific issues accurately and efficiently. The benchmarking process generally takes only a few dozens of minutes to evaluate a tool. We have released our dataset on GitHub as open source, with an aim to better facilitate the community to conduct future work on benchmarking symbolic execution tools.

Estimating covariance matrix from massive high-dimensional and distributed data is significant for various real-world applications. In this paper, we propose a data-aware weighted sampling-based covariance matrix estimator, namely DACE, which can provide an unbiased covariance matrix estimation and attain more accurate estimation under the same compression ratio. Moreover, we extend our proposed DACE to tackle multiclass classification problems with theoretical justification and conduct extensive experiments on both synthetic and real-world data sets to demonstrate the superior performance of our DACE.

Shared memory is widely used for inter-process communication. The shared memory abstraction allows computation to be decoupled from communication, which offers benefits, including portability and ease of programming. To enable shared memory access by processes that are on different machines, distributed shared memory (DSM) can be employed. However, DSM systems can suffer from thrashing: while different processes update certain hot data items, the largest amount of effort is spent on data synchronization, and little progress is made by each process. To avoid interference between processes during data updating while providing shared memory at page granularity, more time is reserved for a writer to hold a page in a traditional manner. In this paper, we report on complex thrashing, which can explain why extending the time of holding a page might not be sufficient to control thrashing. To increase the throughput, we propose a thrashing control mechanism that allows each process to update a set of pages during a period of time, where the pages compose a logical area. Because of the isolation of areas, updates on different areas can be performed concurrently. To allow the areas to be fairly well used, each process is assigned with a random priority for thrashing control. The thrashing control mechanism is implemented on a Linux-based DSM system. Performance results show that the execution time of the applications that are apt to cause system thrashing can be significantly reduced by our approach.

Keyphrase generation (KG) aims to generate a set of keyphrases given a document, which is a fundamental task in natural language processing (NLP). Most previous methods solve this problem in an extractive manner, while recently, several attempts are made under the generative setting using deep neural networks. However, the state-of-the-art generative methods simply treat the document title and the document main body equally, ignoring the leading role of the title to the overall document. To solve this problem, we introduce a new model called Title-Guided Network (TG-Net) for automatic keyphrase generation task based on the encoderdecoder architecture with two new features: (i) the title is additionally employed as a query-like input, and (ii) a titleguided encoder gathers the relevant information from the title to each word in the document. Experiments on a range of KG datasets demonstrate that our model outperforms the state-of-the-art models with a large margin, especially for documents with either very low or very high title length ratios.

We investigate the task of distractor generation for multiple choice reading comprehension questions from examinations. In contrast to all previous works, we do not aim at preparing words or short phrases distractors, instead, we endeavor to generate longer and semantic-rich distractors which are closer to distractors in real reading comprehension from examinations. Taking a reading comprehension article, a pair of question and its correct option as input, our goal is to generate several distractors which are somehow related to the answer, consistent with the semantic context of the question and have some trace in the article. We propose a hierarchical encoderdecoder framework with static and dynamic attention mechanisms to tackle this task. Specifically, the dynamic attention can combine sentence-level and word-level attention varying at each recurrent time step to generate a more readable sequence. The static attention is to modulate the dynamic attention not to focus on question irrelevant sentences or sentences which contribute to the correct option. Our proposed framework outperforms several strong baselines on the first prepared distractor generation dataset of real reading comprehension questions. For human evaluation, compared with those distractors generated by baselines, our generated distractors are more functional to confuse the annotators.

We present DDFlow, a data distillation approach to learning optical flow estimation from unlabeled data. The approach distills reliable predictions from a teacher network, and uses these predictions as annotations to guide a student network to learn optical flow. Unlike existing work relying on handcrafted energy terms to handle occlusion, our approach is data-driven, and learns optical flow for occluded pixels. This enables us to train our model with a much simpler loss function, and achieve a much higher accuracy. We conduct a rigorous evaluation on the challenging Flying Chairs, MPI Sintel, KITTI 2012 and 2015 benchmarks, and show that our approach significantly outperforms all existing unsupervised learning methods, while running at real time.

We study the problem of generating interconnected questions in question-answering style conversations.Compared with previous works which generate questions based on a single sentence (or paragraph), this setting is different in two major aspects: (1) Questions are highly conversational.Almost half of them refer back to conversation history using coreferences.(2) In a coherent conversation, questions have smooth transitions between turns.We propose an end-to-end neural model with coreference alignment and conversation flow modeling.The coreference alignment modeling explicitly aligns coreferent mentions in conversation history with corresponding pronominal references in generated questions, which makes generated questions interconnected to conversation history.The conversation flow modeling builds a coherent conversation by starting questioning on the first few sentences in a text passage and smoothly shifting the focus to later parts.Extensive experiments show that our system outperforms several baselines and can generate highly conversational questions.The code implementation is released at https://github.com/ Evan-Gao/conversaional-QG.

A huge volume of user-generated content is daily produced on social media.To facilitate automatic language understanding, we study keyphrase prediction, distilling salient information from massive posts.While most existing methods extract words from source posts to form keyphrases, we propose a sequence-to-sequence (seq2seq) based neural keyphrase generation framework, enabling absent keyphrases to be created.Moreover, our model, being topic-aware, allows joint modeling of corpus-level latent topic representations, which helps alleviate the data sparsity that widely exhibited in social media language.Experiments on three datasets collected from English and Chinese social media platforms show that our model significantly outperforms both extraction and generation models that do not exploit latent topics. 1 Further discussions show that our model learns meaningful topics, which interprets its superiority in social media keyphrase generation.

We present a self-supervised learning approach for optical flow. Our method distills reliable flow estimations from non-occluded pixels, and uses these predictions as ground truth to learn optical flow for hallucinated occlusions. We further design a simple CNN to utilize temporal information from multiple frames for better flow estimation. These two principles lead to an approach that yields the best performance for unsupervised optical flow learning on the challenging benchmarks including MPI Sintel, KITTI 2012 and 2015. More notably, our self-supervised pre-trained model provides an excellent initialization for supervised fine-tuning. Our fine-tuned models achieve state-of-the-art results on all three datasets. At the time of writing, we achieve EPE=4.26 on the Sintel benchmark, outperforming all submitted methods.

The exceptional performance of Deep neural networks (DNNs) encourages their deployment in safety-and dependability-critical systems. However, DNNs often demonstrate erroneous behaviors in real-world corner cases. Existing countermeasures center on improving the testing and bug-fixing practice. Unfortunately, building a bug-free DNN-based system is almost impossible currently due to its black-box nature, so anomaly detection is imperative in practice. Motivated by the idea of data validation in a traditional program, we propose and implement Deep Validation, a novel framework for detecting real-world error-inducing corner cases in a DNN-based system during runtime. We model the specifications of DNNs by resorting to their training data and cast checking input validity of DNNs as the problem of discrepancy estimation. Deep Validation achieves excellent detection results against various corner case scenarios across three popular datasets. Consequently, Deep Validation greatly complements existing efforts and is a crucial step toward building safe and dependable DNN-based systems.

Shilin He, Zhaopeng Tu, Xing Wang, Longyue Wang, Michael Lyu, Shuming Shi. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Jingjing Li, Yifan Gao, Lidong Bing, Irwin King, Michael R. Lyu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

We investigate the difficulty levels of questions in reading comprehension datasets such as SQuAD, and propose a new question generation setting, named Difficulty-controllable Question Generation (DQG). Taking as input a sentence in the reading comprehension paragraph and some of its text fragments (i.e., answers) that we want to ask questions about, a DQG method needs to generate questions each of which has a given text fragment as its answer, and meanwhile the generation is under the control of specified difficulty labels---the output questions should satisfy the specified difficulty as much as possible. To solve this task, we propose an end-to-end framework to generate questions of designated difficulty levels by exploring a few important intuitions. For evaluation, we prepared the first dataset of reading comprehension questions with difficulty labels. The results show that the question generated by our framework not only have better quality under the metrics like BLEU, but also comply with the specified difficulty labels.

Wasserstein Generative Adversarial Nets~(GANs) are newly proposed GAN algorithms and widely used in computer vision, web mining, information retrieval, etc. However, the existing algorithms with approximated Wasserstein loss converge slowly due to heavy computation cost and usually generate unstable results as well. In this paper, we solve the computation cost problem by speeding up the Wasserstein GANs from a well-designed communication efficient parallel architecture. Specifically, we develop a new problem formulation targeting the accurate evaluation of Wasserstein distance and propose an easily parallel optimization algorithm to train the Wasserstein GANs. Compared to traditional parallel architecture, our proposed framework is designed explicitly for the skew parameter updates between the generator network and discriminator network. Rigorous experiments reveal that our proposed framework achieves a significant improvement regarding convergence speed with comparable stability on generating images, compared to the state-of-the-art of Wasserstein GANs algorithms.

Recent advances in deep learning promote the innovation of many intelligent systems and applications such as autonomous driving and image recognition. Despite enormous efforts and investments in this field, a fundamental question remains under-investigated - what challenges do developers commonly face when building deep learning applications? To seek an answer, this paper presents a large-scale empirical study of deep learning questions in a popular Q&A website, Stack Overflow. We manually inspect a sample of 715 questions and identify seven kinds of frequently asked questions. We further build a classification model to quantify the distribution of different kinds of deep learning questions in the entire set of 39,628 deep learning questions. We find that program crashes, model migration, and implementation questions are the top three most frequently asked questions. After carefully examining accepted answers of these questions, we summarize five main root causes that may deserve attention from the research community, including API misuse, incorrect hyperparameter selection, GPU computation, static graph computation, and limited debugging and profiling support. Our results highlight the need for new techniques such as cross-framework differential testing to improve software development productivity and software reliability in deep learning.

Layout bugs commonly exist in mobile apps. Due to the fragmentation issues of smartphones, a layout bug may occur only on particular versions of smartphones. It is quite challenging to detect such bugs for state-of-the-art commercial automated testing platforms, although they can test an app with thousands of different smartphones in parallel. The main reason is that typical layout bugs neither crash an app nor generate any error messages. In this paper, we present our work for detecting text-layout bugs, which account for a large portion of layout bugs. We model text-layout bug detection as a classification problem. This then allows us to address it with sophisticated image processing and machine learning techniques. To this end, we propose an approach which we call Textout. Textout takes screenshots as its input and adopts a specifically-tailored text detection method and a convolutional neural network (CNN) classifier to perform automatic text-layout bug detection. We collect 33,102 text-region images as our training dataset and verify the effectiveness of our tool with 1,481 text-region images collected from real-world apps. Textout achieves an AUC (area under the curve) of 0.956 on the test dataset and shows an acceptable overhead. The dataset is open-source released for follow-up research.

Previous studies showed that replying to a user review usually has a positive effect on the rating that is given by the user to the app. For example, Hassan et al. found that responding to a review increases the chances of a user updating their given rating by up to six times compared to not responding. To alleviate the labor burden in replying to the bulk of user reviews, developers usually adopt a template-based strategy where the templates can express appreciation for using the app or mention the company email address for users to follow up. However, reading a large number of user reviews every day is not an easy task for developers. Thus, there is a need for more automation to help developers respond to user reviews. Addressing the aforementioned need, in this work we propose a novel approach RRGen that automatically generates review responses by learning knowledge relations between reviews and their responses. RRGen explicitly incorporates review attributes, such as user rating and review length, and learns the relations between reviews and corresponding responses in a supervised way from the available training data. Experiments on 58 apps and 309,246 review-response pairs highlight that RRGen outperforms the baselines by at least 67.4% in terms of BLEU-4 (an accuracy measure that is widely used to evaluate dialogue response generation systems). Qualitative analysis also confirms the effectiveness of RRGen in generating relevant and accurate responses.

System logs record detailed runtime information of software systems and are used as the main data source for many tasks around software engineering. As modern software systems are evolving into large scale and complex structures, logs have become one type of fast-growing big data in industry. In particular, such logs often need to be stored for a long time in practice (e.g., a year), in order to analyze recurrent problems or track security issues. However, archiving logs consumes a large amount of storage space and computing resources, which in turn incurs high operational cost. Data compression is essential to reduce the cost of log storage. Traditional compression tools (e.g., gzip) work well for general texts, but are not tailed for system logs. In this paper, we propose a novel and effective log compression method, namely logzip. Logzip is capable of extracting hidden structures from raw logs via fast iterative clustering and further generating coherent intermediate representations that allow for more effective compression. We evaluate logzip on five large log datasets of different system types, with a total of 63.6 GB in size. The results show that logzip can save about half of the storage space on average over traditional compression tools. Meanwhile, the design of logzip is highly parallel and only incurs negligible overhead. In addition, we share our industrial experience of applying logzip to Huawei's real products.

This paper presents an unsupervised framework for jointly modeling topic content and discourse behavior in microblog conversations. Concretely, we propose a neural model to discover word clusters indicating what a conversation concerns (i.e., topics) and those reflecting how participants voice their opinions (i.e., discourse). 1 Extensive experiments show that our model can yield both coherent topics and meaningful discourse behavior. Further study shows that our topic and discourse representations can benefit the classification of microblog messages, especially when they are jointly trained with the classifier. Our data sets and code are available at: http://github.com/zengjichuan/Topic_Disc .

Topic modelling and citation recommendation of scientific articles are important yet challenging research problems in scientific article analysis. In particular, the inference on coherent topics can be easily affected by irrelevant contents in articles. Meanwhile, the extreme sparsity of citation networks brings difficulty to a valid citation recommendation. Intuitively, articles with similar topics are more likely to cite each other, and cited articles tend to share similar themes. Motivated from this intuition, we aim to boost the performance of both topic modelling and citation recommendation by effectively leverage this underlying correlation between latent topics and citation networks. To this end, we propose a novel Bayesian deep generative model termed as Neural Relational Topic Model (NRTM), which is composed with a Stacked Variational Auto-Encoder (SVAE) and a multilayer perception (MLP). Specifically, the SVAE utilizes an inference network to learn more representative topics of document contents, which can help to enrich the latent factors in collaborative filtering of citations. Furthermore, the MLP network conducts nonlinear collaborative filtering of citations, which can further benefit the inference of topics by leveraging the knowledge of citation networks. Extensive experiments on two real-world datasets demonstrate that our model can effectively take advantages of the coherence between topic learning and citation recommendation, and significantly outperform the state-of-the-art methods on both tasks.

Deep metric learning is widely used in extreme classification and image retrieval because of its powerful ability to learn the semantic low-dimensional embedding of high-dimensional data. However, the heavy computational cost of mining valuable pair or triplet of training data and updating models frequently in existing deep metric learning approaches becomes a barrier to apply such methods to a large-scale real-world context in a distributed environment. Moreover, existing distributed deep learning framework is not designed for deep metric learning tasks, because it is difficult to implement a smart mining policy of valuable training data. In this paper, we introduce a novel distributed framework to speed up the training process of the deep metric learning using multiple machines. Specifically, we first design a distributed sampling method to find the hard-negative samples from a broader scope of candidate samples compared to the single-machine solution. Then, we design a hybrid communication pattern and implement a decentralized data-parallel framework to reduce the communication workload while the quality of the trained deep metric models is preserved. In experiments, we show excellent performance gain compared to a full spectrum of state-of-the-art deep metric learning models on multiple datasets in terms of image clustering and image retrieval tasks.

Control-flow obfuscation increases program complexity by semantic-preserving transformation. Opaque predicates are essential gadgets to achieve such transformation. However, we observe that real-world opaque predicates are generally very simple and engage little security consideration. Recently, such insecure opaque predicates have been severely attacked by symbolic execution-based adversaries and jeopardize the security of control-flow obfuscation. This paper, therefore, proposes symbolic opaque predicates which can be resilient to symbolic execution-based adversaries. We design a general framework to compose such opaque predicates, which requires introducing challenging symbolic analysis problems (e.g., symbolic memory) in each opaque predicate. In this way, we may mislead symbolic execution engines into reaching false conclusions. We observe a novel bi-opaque property about symbolic opaque predicates, which can incur not only false negative issues but also false positive issues to attackers. To evaluate the efficacy of our idea, we have implemented a prototype obfuscation tool based on Obfuscator-LLVM and conduct experiments with real-world programs. Our evaluation results show that symbolic opaque predicates demonstrate excellent resilience to prevalent symbolic execution engines, such as BAP, Triton, and Angr. Moreover, although the costs of symbolic opaque predicates may vary for different problem settings, some predicates can be very efficient. Therefore, our framework is both secure and usable. Users can follow the framework to introduce symbolic opaque predicates into their obfuscation tools and made them more powerful.

Multi-head attention is appealing for the ability to jointly attend to information from different representation subspaces at different positions.In this work, we introduce a disagreement regularization to explicitly encourage the diversity among multiple attention heads.Specifically, we propose three types of disagreement regularization, which respectively encourage the subspace, the attended positions, and the output representation associated with each attention head to be different from other heads.Experimental results on widely-used WMT14 English⇒German and WMT17 Chinese⇒English translation tasks demonstrate the effectiveness and universality of the proposed approach.* Zhaopeng Tu is the corresponding author of the paper.This work was mainly conducted when Jian Li and Baosong Yang were interning at Tencent AI Lab.

Many classification models work poorly on short texts due to data sparsity.To address this issue, we propose topic memory networks for short text classification with a novel topic memory mechanism to encode latent topic representations indicative of class labels.Different from most prior work that focuses on extending features with external knowledge or pre-trained topics, our model jointly explores topic inference and text classification with memory networks in an end-to-end manner.Experimental results on four benchmark datasets show that our model outperforms state-of-the-art models on short text classification, meanwhile generates coherent topics.* This work was mainly conducted when Jichuan Zeng was an intern in Tencent AI Lab.† Jing Li is the corresponding author.Training instances R1: [SuperBowl] I'll do anything to see the Steelers win.R2: [New.Music.Live] Please give wristbands, she have major Bieber Fever.

Check-in prediction is an important task for location-based systems, which maps a noisy estimate of a user's current location to a semantically meaningful point-of-interest (POI), such as a restaurant or store. In this paper, we leverage the personalized preference and sequential check-in pattern to improve the traditional methods that base on the geographical and temporal contexts. In our approach, we propose a Gaussian mixture model and a histogram distribution estimation model to learn the contextual features from relevant spatial and temporal information, respectively. Furthermore, we employ user and POI embeddings to model the personalized preference and leverage a stacked Long-Short Term Memory (LSTM) model to learn the sequential check-in pattern. Combining the contextual features and the personalized sequential patterns together, we propose a wide and deep neural network for the check-in prediction task. Experimental evaluations on two real-life datasets demonstrate that our proposed method outperforms state-of-the-art models.

Detecting emerging issues (e.g., new bugs) timely and precisely is crucial for developers to update their apps. App reviews provide an opportunity to proactively collect user complaints and promptly improve apps' user experience, in terms of bug fixing and feature refinement. However, the tremendous quantities of reviews and noise words (e.g., misspelled words) increase the difficulties in accurately identifying newly-appearing app issues. In this paper, we propose a novel and automated framework IDEA, which aims to IDentify Emerging App issues effectively based on online review analysis. We evaluate IDEA on six popular apps from Google Play and Apple's App Store, employing the official app changelogs as our ground truth. Experiment results demonstrate the effectiveness of IDEA in identifying emerging app issues. Feedback from engineers and product managers shows that 88.9% of them think that the identified issues can facilitate app development in practice. Moreover, we have successfully applied IDEA to several products of Tencent, which serve hundreds of millions of users.

Intelligent code completion has become an essential research task to accelerate modern software development. To facilitate effective code completion for dynamically-typed programming languages, we apply neural language models by learning from large codebases, and develop a tailored attention mechanism for code completion. However, standard neural language models even with attention mechanism cannot correctly predict the out-of-vocabulary (OoV) words that restrict the code completion performance. In this paper, inspired by the prevalence of locally repeated terms in program source code, and the recently proposed pointer copy mechanism, we propose a pointer mixture network for better predicting OoV words in code completion. Based on the context, the pointer mixture network learns to either generate a within-vocabulary word through an RNN component, or regenerate an OoV word from local context through a pointer component. Experiments on two benchmarked datasets demonstrate the effectiveness of our attention mechanism and pointer mixture network on the code completion task.

In this paper, we propose a generic approach for accelerating the convergence of existing algorithms to solve the problem of stochastic zeroth-order convex optimization (SZCO). Standard techniques for accelerating the convergence of stochastic zeroth-order algorithms are by exploring multiple functional evaluations (e.g., two-point evaluations), or by exploiting global conditions of the problem (e.g., smoothness and strong convexity). Nevertheless, these classic acceleration techniques are necessarily restricting the applicability of newly developed algorithms. The key of our proposed generic approach is to explore a local growth condition (or called local error bound condition) of the objective function in SZCO. The benefits of the proposed acceleration technique are: (i) it is applicable to both settings with one-point evaluation and two-point evaluations; (ii) it does not necessarily require strong convexity or smoothness condition of the objective function; (iii) it yields an improvement on convergence for a broad family of problems. Empirical studies in various settings demonstrate the effectiveness of the proposed acceleration approach.

Logging is a common programming practice of great importance in modern software development, because software logs have been widely used in various software maintenance tasks. To provide high-quality logs, developers need to design the description text in logging statements carefully. Inappropriate descriptions will slow down or even mislead the maintenance process, such as postmortem analysis. However, there is currently a lack of rigorous guide and specifications on developer logging behaviors, which makes the construction of description text in logging statements a challenging problem. To fill this significant gap, in this paper, we systematically study what developers log, with focus on the usage of natural language descriptions in logging statements. We obtain 6 valuable findings by conducting source code analysis on 10 Java projects and 7 C# projects, which contain 28,532,975 LOC and 115,159 logging statements in total. Furthermore, our study demonstrates the potential of automated description text generation for logging statements by obtaining up to 49.04 BLEU-4 score and 62.1 ROUGE-L score using a simple information retrieval method. To facilitate future research in this field, the datasets have been publicly released.

In linear stochastic bandits, it is commonly assumed that payoffs are with sub-Gaussian noises. In this paper, under a weaker assumption on noises, we study the problem of \underline{lin}ear stochastic {\underline b}andits with h{\underline e}avy-{\underline t}ailed payoffs (LinBET), where the distributions have finite moments of order $1+\epsilon$, for some $\epsilon\in (0,1]$. We rigorously analyze the regret lower bound of LinBET as $\Omega(T^{\frac{1}{1+\epsilon}})$, implying that finite moments of order 2 (i.e., finite variances) yield the bound of $\Omega(\sqrt{T})$, with $T$ being the total number of rounds to play bandits. The provided lower bound also indicates that the state-of-the-art algorithms for LinBET are far from optimal. By adopting median of means with a well-designed allocation of decisions and truncation based on historical information, we develop two novel bandit algorithms, where the regret upper bounds match the lower bound up to polylogarithmic factors. To the best of our knowledge, we are the first to solve LinBET optimally in the sense of the polynomial order on $T$. Our proposed algorithms are evaluated based on synthetic datasets, and outperform the state-of-the-art results.

App reviews play an essential role for users to convey their feedback about using the app. The critical information contained in app reviews can assist app developers for maintaining and updating mobile apps. However, the noisy nature and large-quantity of daily generated app reviews make it difficult to understand essential information carried in app reviews. Several prior studies have proposed methods that can automatically classify or cluster user reviews into a few app topics (e.g., security). These methods usually act on a static collection of user reviews. However, due to the dynamic nature of user feedback (i.e., reviews keep coming as new users register or new app versions being released) and multiple analysis dimensions (e.g., review quantity and user rating), developers still need to spend substantial effort in extracting contrastive information that can only be teased out by comparing data from multiple time periods or analysis dimensions. This is needed to answer questions such as: what kind of issues users are experiencing most? is there an unexpected rise in a particular kind of issue? etc. To address this need, in this paper, we introduce INFAR, a tool that automatically extracts INsights From App Reviews across time periods and analysis dimensions, and presents them in natural language supported by an interactive chart. The insights INFAR extracts include several perspectives: (1) salient topics (i.e., issue topics with significantly lower ratings), (2) abnormal topics (i.e., issue topics that experience a rapid rise in volume during a time period), (3) correlations between two topics, and (4) causal factors to rating or review quantity changes. To evaluate our tool, we conduct an empirical evaluation by involving six popular apps and 12 industrial practitioners, and 92% (11/12) of them approve the practical usefulness of the insights summarized by INFAR.

Logs are often used for troubleshooting in large-scale software systems. For a cloud-based online system that provides 24/7 service, a huge number of logs could be generated every day. However, these logs are highly imbalanced in general, because most logs indicate normal system operations, and only a small percentage of logs reveal impactful problems. Problems that lead to the decline of system KPIs (Key Performance Indicators) are impactful and should be fixed by engineers with a high priority. Furthermore, there are various types of system problems, which are hard to be distinguished manually. In this paper, we propose Log3C, a novel clustering-based approach to promptly and precisely identify impactful system problems, by utilizing both log sequences (a sequence of log events) and system KPIs. More specifically, we design a novel cascading clustering algorithm, which can greatly save the clustering time while keeping high accuracy by iteratively sampling, clustering, and matching log sequences. We then identify the impactful problems by correlating the clusters of log sequences with system KPIs. Log3C is evaluated on real-world log data collected from an online service system at Microsoft, and the results confirm its effectiveness and efficiency. Furthermore, our approach has been successfully applied in industrial practice.

In the next decade, Artificial Intelligent (AI) techniques can see wide adoption in our daily life to release human burden. In our recent Software Engineering research, we investigated on the design of novel AI methods to facilitate all three major phases in software engineering: development, operation, and analysis. In this talk, I will first introduce the AI techniques we employed, including machine learning framework, classification, clustering, matrix factorization, topic modeling, deep learning, and parallel computing platform. Then I will explain the challenges in each phase and describe our recently proposed methodologies. First in development phase, we suggested an automated code completion technique via deep learning. Our technique learns the code style from lots of existing code bases, and recommends the most suitable token based on the trained deep learning model and current coding context. Besides, to help developers in conducting effective logging, we designed a tool named LogAdvisor, which tells developers whether they should write a logging statement in the current code block or not. Secondly, in operation phase, we implemented a continuous and passive authentication method for mobile phones based on user touch biometrics. Different from the traditional password authentication scheme, our method can recognize malicious attackers based on abnormal user behaviors. Moreover, we developed PAID, which automatically prioritizes app issues by mining user reviews. Finally, in analysis phase, we designed systematic data analytics techniques for software reliability prediction. Besides, to make full use of the crucial runtime information, we proposed effective methods for every step in log analysis, including log parsing, feature extraction, and log mining. Furthermore, we developed a CNN-based defect prediction method to help developers find the buggy code. In the end, we expect to establish a comprehensive framework for systematic employment of AI techniques in the Software Engineering paradigm.

Even well-designed software systems suffer from chronic performance degradation, also known as “software aging”, due to internal (e.g., software bugs) or external (e.g., resource exhaustion) impairments. These chronic problems often fly under the radar of software monitoring systems before causing severe impacts (e.g., system failures). Therefore, it is a challenging issue how to timely predict the occurrence of failures caused by these problems. Unfortunately, the effectiveness of prior approaches are far from satisfactory due to the insufficiency of aging indicators adopted by them. To accurately predict failures caused by software aging which are named as Aging-Related Failure (ARFs), this paper presents a novel entropy-based aging indicator, namely Multidimensional Multi-scale Entropy (MMSE) which leverages the complexity embedded in runtime performance metrics to indicate software aging. To the best of our knowledge, this is the first time to leverage entropy to predict ARFs. Based upon MMSE, we implement three failure prediction approaches encapsulated in a proof-of-concept prototype named ARF-Predictor. The experimental evaluations in a Video on Demand (VoD) system, and in a real-world production system, AntVision, show that ARF-Predictor can predict ARFs with a very high accuracy and a low Ahead-Time-To-Failure (ATTF). Compared to previous approaches, ARF-Predictor improves the prediction accuracy by about 5 times and reduces ATTF even by 3 orders of magnitude. In addition, ARF-Predictor is light-weight enough to satisfy the real-time requirement.

Logs are widely used in system management for dependability assurance because they are often the only data available that record detailed system runtime behaviors in production. Because the size of logs is constantly increasing, developers (and operators) intend to automate their analysis by applying data mining methods, therefore structured input data (e.g., matrices) are required. This triggers a number of studies on log parsing that aims to transform free-text log messages into structured events. However, due to the lack of open-source implementations of these log parsers and benchmarks for performance comparison, developers are unlikely to be aware of the effectiveness of existing log parsers and their limitations when applying them into practice. They must often reimplement or redesign one, which is time-consuming and redundant. In this paper, we first present a characterization study of the current state of the art log parsers and evaluate their efficacy on five real-world datasets with over ten million log messages. We determine that, although the overall accuracy of these parsers is high, they are not robust across all datasets. When logs grow to a large scale (e.g., 200 million log messages), which is common in practice, these parsers are not efficient enough to handle such data on a single computer. To address the above limitations, we design and implement a parallel log parser (namely POP) on top of Spark, a large-scale data processing platform. Comprehensive experiments have been conducted to evaluate POP on both synthetic and real-world datasets. The evaluation results demonstrate the capability of POP in terms of accuracy, efficiency, and effectiveness on subsequent log mining tasks.

Classifying binary imbalanced streaming data is a significant task in both machine learning and data mining. Previously, online area under the receiver operating characteristic (ROC) curve (AUC) maximization has been proposed to seek a linear classifier. However, it is not well suited for handling nonlinearity and heterogeneity of the data. In this paper, we propose the kernelized online imbalanced learning (KOIL) algorithm, which produces a nonlinear classifier for the data by maximizing the AUC score while minimizing a functional regularizer. We address four major challenges that arise from our approach. First, to control the number of support vectors without sacrificing the model performance, we introduce two buffers with fixed budgets to capture the global information on the decision boundary by storing the corresponding learned support vectors. Second, to restrict the fluctuation of the learned decision function and achieve smooth updating, we confine the influence on a new support vector to its -nearest opposite support vectors. Third, to avoid information loss, we propose an effective compensation scheme after the replacement is conducted when either buffer is full. With such a compensation scheme, the performance of the learned model is comparable to the one learned with infinite budgets. Fourth, to determine good kernels for data similarity representation, we exploit the multiple kernel learning framework to automatically learn a set of kernels. Extensive experiments on both synthetic and real-world benchmark data sets demonstrate the efficacy of our proposed approach.

The dramatic rise of Android application (app) marketplaces has significantly gained the success of convenience for mobile users. Consequently, with the advantage of numerous Android apps, Android malware seizes the opportunity to steal privacy-sensitive data by pretending to provide functionalities as benign apps do. To distinguish malware from millions of Android apps, researchers have proposed sophisticated static and dynamic analysis tools to automatically detect and classify malicious apps. Most of these tools, however, rely on manual configuration of lists of features based on permissions, sensitive resources, intents, etc., which are difficult to come by. To address this problem, we study real-world Android apps to mine hidden patterns of malware and are able to extract highly sensitive APIs that are widely used in Android malware. We also implement an automated malware detection system, MalPat, to fight against malware and assist Android app marketplaces to address unknown malicious apps. Comprehensive experiments are conducted on our dataset consisting of 31 185 benign apps and 15 336 malware samples. Experimental results show that MalPat is capable of detecting malware with a high $F_1$ score (98.24%) comparing with the state-of-the-art approaches.

User request trace-oriented monitoring is an effective method to improve the reliability of cloud services. However, there are some difficulties in getting useful traces in practice, which hinder the development of trace-oriented monitoring research. In this paper, we release a fine-grained user request-centric open trace data set, called TraceBench, which is collected in a real-world cloud storage service deployed in a real environment. When collecting, we consider different scenarios, involving multiple scales of clusters, different kinds of user requests, various speeds of workloads, many types of injected faults, etc. To validate the usability and authenticity, we have employed TraceBench in several trace-oriented monitoring topics, such as anomaly detection, performance problem diagnosis, and temporal invariant mining. The results show that TraceBench well supports these research topics. In addition, we have also carried out an extensive data analysis based on TraceBench, which validates the high quality of the data set.

Contextual bandits with linear payoffs, which are also known as linear bandits, provide a powerful alternative for solving practical problems of sequential decisions, e.g., online advertisements. In the era of big data, contextual data usually tend to be high-dimensional, which leads to new challenges for traditional linear bandits mostly designed for the setting of low-dimensional contextual data. Due to the curse of dimensionality, there are two challenges in most of the current bandit algorithms: the first is high time-complexity; and the second is extreme large upper regret bounds with high-dimensional data. In this paper, in order to attack the above two challenges effectively, we develop an algorithm of Contextual Bandits via RAndom Projection (CBRAP) in the setting of linear payoffs, which works especially for high-dimensional contextual data. The proposed CBRAP algorithm is time-efficient and flexible, because it enables players to choose an arm in a low-dimensional space, and relaxes the sparsity assumption of constant number of non-zero components in previous work. Besides, we prove an upper regret bound for the proposed algorithm, which is associated with reduced dimensions. By comparing with three benchmark algorithms, we demonstrate improved performance on cumulative payoffs of CBRAP during its sequential decisions on both synthetic and real-world datasets, as well as its superior time-efficiency.

Concolic execution has achieved great success in many binary analysis tasks. However, it is still not a primary option for industrial usage. A well-known reason is that concolic execution cannot scale up to large-size programs. Many research efforts have focused on improving its scalability. Nonetheless, we find that, even when processing small-size programs, concolic execution suffers a great deal from the accuracy and scalability issues. This paper systematically investigates the challenges that can be introduced even by small-size programs, such as symbolic array and symbolic jump. We further verify that the proposed challenges are non-trivial via real-world experiments with three most popular concolic execution tools: BAP, Triton, and Angr. Among a set of 22 logic bombs we designed, Angr can solve only four cases correctly, while BAP and Triton perform much worse. The results imply that current tools are still primitive for practical industrial usage. We summarize the reasons and release the bombs as open source to facilitate further study.

In smart cities, green cellular networks play a crucial role to support wireless access for numerous devices anywhere and anytime with efficiency and sustainability. Because base stations (BSes) consume more than 70% of overall cellular network infrastructure energy, saving the power consumption of BSes is the key task to build a green cellular network. Except for low power design of the BS hardware and software, the traffic-driven BS sleeping operation is an economical way to improve existing cellular networks, which can reduce the BS power consumption at low traffic load. However, prior BS sleeping strategies establish on the static temporal characteristics of traffic load, which ignore the fact that network traffic is influenced by many factors such as time, human mobility, holiday, weather, etc. Hence, prior traffic estimation is coarse, and the BS sleeping strategies cannot apply to the changing network traffic. In this paper, we exploit a machine learning method to estimate the BS traffic and propose a BS sleeping strategy based on predicted traffic for power saving in the cellular network. We analyze network traffic in multi-views: temporal influence, spatial influence, and event influence. Then, we propose a multi-view ensemble learning model to predict network traffic load, which learns the traffic in multi-views and combine the results with ensemble. Furthermore, we formulate a BS sleeping strategy based on the predicted traffic load. Finally, we evaluate our traffic prediction algorithm on real cellular network data. The evaluation shows that our traffic prediction algorithm improves about 40% than state-of-the-art machine learning methods. Also, we evaluate the proposed BS sleeping strategy, which yields about 10% more energy savings and less device damage than the competitors in the simulated environment.

Best arm identification in stochastic Multi-Armed Bandits (MAB) has become an essential variant in the research line of bandits for decision-making problems. In previous work, the best arm usually refers to an arm with the highest expected payoff in a given decision-arm set. However, in many practical scenarios, it would be more important and desirable to incorporate the risk of an arm into the best decision. In this paper, motivated by practical applications with risk via bandits, we investigate the problem of Risk Control of Best Arm Identification (RCBAI) in stochastic MAB. Based on the technique of Successive Rejects (SR), we show that the error resulting from the mean-variance estimation is sub-Gamma by setting mild assumptions on stochastic payoffs of arms. Besides, we develop an algorithm named as RCMAB. SR, and derive an upper bound for the probability of error for RCBAI in stochastic MAB. We demonstrate the superiority of the RCMAB. SR algorithm in synthetic datasets, and then apply the RCMAB. SR algorithm in financial data for yearly investments to show its superiority for practical applications.

Estimating covariance matrices is a fundamental technique in various domains, most notably in machine learning and signal processing. To tackle the challenges of extensive communication costs, large storage capacity requirements, and high processing time complexity when handling massive high-dimensional and distributed data, we propose an efficient and accurate covariance matrix estimation method via data compression. In contrast to previous data-oblivious compression schemes, we leverage a data-aware weighted sampling method to construct low-dimensional data for such estimation. We rigorously prove that our proposed estimator is unbiased and requires smaller data to achieve the same accuracy with specially designed sampling distributions. Besides, we depict that the computational procedures in our algorithm are efficient. All achievements imply an improved tradeoff between the estimation accuracy and computational costs. Finally, the extensive experiments on synthetic and real-world datasets validate the superior property of our method and illustrate that it significantly outperforms the state-of-the-art algorithms.

Logs, which record valuable system runtime information, have been widely employed in Web service management by service providers and users. A typical log analysis based Web service management procedure is to first parse raw log messages because of their unstructured format; and then apply data mining models to extract critical system behavior information, which can assist Web service management. Most of the existing log parsing methods focus on offline, batch processing of logs. However, as the volume of logs increases rapidly, model training of offline log parsing methods, which employs all existing logs after log collection, becomes time consuming. To address this problem, we propose an online log parsing method, namely Drain, that can parse logs in a streaming and timely manner. To accelerate the parsing process, Drain uses a fixed depth parse tree, which encodes specially designed rules for parsing. We evaluate Drain on five real-world log data sets with more than 10 million raw log messages. The experimental results show that Drain has the highest accuracy on four data sets, and comparable accuracy on the remaining one. Besides, Drain obtains 51.85%~81.47% improvement in running time compared with the state-of-the-art online parser. We also conduct a case study on an anomaly detection task using Drain in the parsing step, which determines the effectiveness of Drain in log analysis.

Reliability prediction is an important task in software reliability engineering, which has been widely studied in the last decades. However, modelling and predicting user-perceived reliability of black-box services remain an open research problem. Software services, such as Web services and Web APIs, generally provide black-box functionalities to users through the Internet, thus leading to a lack of their internal information for reliability analysis. Furthermore, the user-perceived service reliability depends not only on the service itself, but also heavily on the invocation context (e.g., service workloads, network conditions), whereby traditional reliability models become ineffective and inappropriate. To address these new challenges posed by blackbox services, in this paper, we propose CARP, a new contextaware reliability prediction approach, which leverages historical usage data from users to construct context-aware reliability models and further provides online reliability prediction results to users. Through context-aware reliability modelling, CARP is able to alleviate the data sparsity problem that heavily limits the prediction accuracy of other existing approaches. The preliminary evaluation results show that CARP can make a significant improvement in reliability prediction accuracy, e.g., about 41% in MAE and 38% in RMSE when only 5% of the data are available.

Probabilistic Temporal Tensor Factorization (PTTF) is an effective algorithm to model the temporal tensor data. It leverages a time constraint to capture the evolving properties of tensor data. Nowadays the exploding dataset demands a large scale PTTF analysis, and a parallel solution is critical to accommodate the trend. Whereas, the parallelization of PTTF still remains unexplored. In this paper, we propose a simple yet efficient Parallel Probabilistic Temporal Tensor Factorization, referred to as P2T2F, to provide a scalable PTTF solution. P2T2F is fundamentally disparate from existing parallel tensor factorizations by considering the probabilistic decomposition and the temporal effects of tensor data. It adopts a new tensor data split strategy to subdivide a large tensor into independent sub-tensors, the computation of which is inherently parallel. We train P2T2F with an efficient algorithm of stochastic Alternating Direction Method of Multipliers, and show that the convergence is guaranteed. Experiments on several real-word tensor datasets demonstrate that P2T2F is a highly effective and efficiently scalable algorithm dedicated for large scale probabilistic temporal tensor analysis.

To improve software reliability, software defect prediction is utilized to assist developers in finding potential bugs and allocating their testing efforts. Traditional defect prediction studies mainly focus on designing hand-crafted features, which are input into machine learning classifiers to identify defective code. However, these hand-crafted features often fail to capture the semantic and structural information of programs. Such information is important in modeling program functionality and can lead to more accurate defect prediction. In this paper, we propose a framework called Defect Prediction via Convolutional Neural Network (DP-CNN), which leverages deep learning for effective feature generation. Specifically, based on the programs' Abstract Syntax Trees (ASTs), we first extract token vectors, which are then encoded as numerical vectors via mapping and word embedding. We feed the numerical vectors into Convolutional Neural Network to automatically learn semantic and structural features of programs. After that, we combine the learned features with traditional hand-crafted features, for accurate software defect prediction. We evaluate our method on seven open source projects in terms of F-measure in defect prediction. The experimental results show that in average, DP-CNN improves the state-of-the-art method by 12%.

Many learning-to-rank~(LtR) algorithms focus on query-independent model, in which query and document do not lie in the same feature space, and the rankers rely on the feature ensemble about query-document pair instead of the similarity between query instance and documents. However, existing algorithms do not consider local structures in query-document feature space, and are fragile to irrelevant noise features. In this paper, we propose a novel Riemannian metric learning algorithm to capture the local structures and develop a robust LtR algorithm. First, we design a concept called ideal candidate document to introduce metric learning algorithm to query-independent model. Previous metric learning algorithms aiming to find an optimal metric space are only suitable for query-dependent model, in which query instance and documents belong to the same feature space and the similarity is directly computed from the metric space. Then we extend the new and extremely fast global Geometric Mean Metric Learning (GMML) algorithm to develop a localized GMML, namely L-GMML. Based on the combination of local learned metrics, we employ the popular Normalized Discounted Cumulative Gain~(NDCG) scorer and Weighted Approximate Rank Pairwise~(WARP) loss to optimize the ideal candidate document for each query candidate set. Finally, we can quickly evaluate all candidates via the similarity between the ideal candidate document and other candidates. By leveraging the ability of metric learning algorithms to describe the complex structural information, our approach gives us a principled and efficient way to perform LtR tasks. The experiments on real-world datasets demonstrate that our proposed L-GMML algorithm outperforms the state-of-the-art metric learning to rank methods and the stylish query-independent LtR algorithms regarding accuracy and computational efficiency.

Urbanization's rapid progress has modernized a large number of human beings' lives. This urbanization progress is accompanied by the increase of a variety of shops (e.g., restaurants and fitness centers) to meet the increasing citizens, which means business opportunities for the investors. Nevertheless, it is difficult for the investors to catch such opportunities because opening what kind of business at which place is not easy to decide. In this paper, we take this challenge and define the business opportunity mining problem, which recommends new business categories at a partitioned business district. Specifically, we exploit the data from location-based social networks (LBSNs) to mine the business opportunities, guiding the business owners to open new commercial shops in certain categories at a particular area. First, we define the properties of a business district and propose a greedy algorithm to partition a city into different districts. Next, we propose an embedding model to learn latent representations of categories, which captures the functional correlations among business categories. Furthermore, we propose a ranking model based on the pairwise loss to recommend categories for a specific district. Finally, we conduct experiments on Yelp data, and experimental results show that our proposed method outperforms the baseline methods and resolves the problem well.

Cloud applications built on service-oriented architectures generally integrate a number of component services to fulfill certain application logic. The changing cloud environment highlights the need for these applications to keep resilient against QoS variations of their component services so that end-to-end quality-of-service (QoS) can be guaranteed. Runtime service adaptation is a key technique to achieve this goal. To support timely and accurate adaptation decisions, effective and efficient QoS prediction is needed to obtain real-time QoS information of component services. However, current research has focused mostly on QoS prediction of working services that are being used by a cloud application, but little on predicting QoS values of candidate services that are equally important in determining optimal adaptation actions. In this paper, we propose an adaptive matrix factorization (namely AMF) approach to perform online QoS prediction for candidate services. AMF is inspired from the widely-used collaborative filtering techniques in recommender systems, but significantly extends the conventional matrix factorization model with new techniques of data transformation, online learning, and adaptive weights. Comprehensive experiments, as well as a case study, have been conducted based on a real-world QoS dataset of Web services (with over 40 million QoS records). The evaluation results demonstrate AMF's superiority in achieving accuracy, efficiency, and robustness, which are essential to enable optimal runtime service adaptation.

With rapid adoption of the cloud computing model, many enterprises have begun deploying cloud-based services. Failures of virtual machines (VMs) in clouds have caused serious quality assurance issues for those services. VM replication is a commonly used technique for enhancing the reliability of cloud services. However, when determining the VM redundancy strategy for a specific service, many state-of-the-art methods ignore the huge network resource consumption issue that could be experienced when the service is in failure recovery mode. This paper proposes a redundant VM placement optimization approach to enhancing the reliability of cloud services. The approach employs three algorithms. The first algorithm selects an appropriate set of VM-hosting servers from a potentially large set of candidate host servers based upon the network topology. The second algorithm determines an optimal strategy to place the primary and backup VMs on the selected host servers with k-fault-tolerance assurance. Lastly, a heuristic is used to address the task-to-VM reassignment optimization problem, which is formulated as finding a maximum weight matching in bipartite graphs. The evaluation results show that the proposed approach outperforms four other representative methods in network resource consumption in the service recovery stage.

A service-oriented System of Systems (SoS) considers a system as a service and constructs a robust and value-added SoS by outsourcing external component systems through service composition techniques. Online reliability prediction for the component systems for the purpose of assuring the overall Quality of Service (QoS) is often a major challenge in coping with a loosely coupled SoS operating under dynamic and uncertain running environments. It is also a prerequisite for guaranteeing runtime QoS of a SoS through optimal service selection for reliable system construction. We propose a novel online reliability time series prediction approach for the component systems in a service-oriented SoS. We utilize Probabilistic Graphical Models (PGMs) to yield near-future, time series predictions. We assess the approach via invocation records collected from widely used real Web services. Experimental results have confirmed the effectiveness of the approach.

Successive point-of-interest (POI) recommendation in location-based social networks (LBSNs) becomes a significant task since it helps users to navigate a number of candidate POIs and provides the best POI recommendations based on users’ most recent check-in knowledge. However, all existing methods for successive POI recommendation only focus on modeling the correlation between POIs based on users’ check-in sequences, but ignore an important fact that successive POI recommendation is a time-subtle recommendation task. In fact, even with the same previous check-in information, users would prefer different successive POIs at different time. To capture the impact of time on successive POI recommendation, in this paper, we propose a spatial-temporal latent ranking (STELLAR) method to explicitly model the interactions among user, POI, and time. In particular, the proposed STELLAR model is built upon a ranking-based pairwise tensor factorization framework with a fine-grained modeling of user-POI, POI-time, and POI-POI interactions for successive POI recommendation. Moreover, we propose a new interval-aware weight utility function to differentiate successive check-ins’ correlations, which breaks the time interval constraint in prior work. Evaluations on two real-world datasets demonstrate that the STELLAR model outperforms state-of-the-art successive POI recommendation model about 20% in Precision@5 and Recall@5.

Logs, which record runtime information of modern systems, are widely utilized by developers (and operators) in system development and maintenance. Due to the ever-increasing size of logs, data mining models are often adopted to help developers extract system behavior information. However, before feeding logs into data mining models, logs need to be parsed by a log parser because of their unstructured format. Although log parsing has been widely studied in recent years, users are still unaware of the advantages of different log parsers nor the impact of them on subsequent log mining tasks. Thus they often re-implement or even re-design a new log parser, which would be time-consuming yet redundant. To address this issue, in this paper, we study four log parsers and package them into a toolkit to allow their reuse. In addition, we obtain six insightful findings by evaluating the performance of the log parsers on five datasets with over ten million raw log messages, while their effectiveness on a real-world log mining task has been thoroughly examined.

Web service recommendation has recently drawn much attention with the growing amount of Web services. Previous work usually exploits the collaborative filtering techniques for Web service recommendation, but suffers from the data sparsity problem that leads to inaccurate results. Our analysis on a real-world Quality of Service (QoS) dataset shows that there is a hidden correlation among users and services. We define such hidden correlation with an asymmetric matrix (namely asymmetric correlation), in which each entry presents the hidden correlation between a user pair or between a service pair. The goal of this work is to employ such asymmetric correlation among users and services to alleviate the data sparsity problem and further enhance the prediction accuracy in service recommendation. Specifically, we propose an asymmetric correlation regularized matrix factorization (MF) framework, in which asymmetric correlation and asymmetric correlation propagation have been naturally integrated. Finally, experimental results on a well-known real-world QoS dataset validate that the use of asymmetric correlation among users and services is effective in improving prediction accuracy for Web service recommendation.

Distance metric learning (DML) is an effective similarity learning tool to learn a distance function from examples to enhance the model performance in applications of classification, regression, and ranking, etc. Most DML algorithms need to learn a Mahalanobis matrix, a positive semidefinite matrix that scales quadratically with the number of dimensions of input data. This brings huge computational cost in the learning procedure, and makes all proposed algorithms infeasible for extremely high-dimensional data even with the low-rank approximation. Differently, in this paper, we take advantage of the power of parallel computation and propose a novel distributed distance metric learning algorithm based on a state-of-the-art DML algorithm, Information-Theoretic Metric Learning (ITML).More specifically, we utilize the property that each positive semidefinite matrix can be decomposed into a combination of rank-one and trace-one matrices and convert the original sequential training procedure into a parallel one. In most cases, the communication demands of the proposed method are also reduced from O(d2) to O(cd), where d is the number of dimensions of the data and c is the number of constraints in DML and can be smaller than d by appropriate selection. Moreover importantly, we present a rigorous theoretical analysis to upper bound the Bregman divergence between the sequential algorithm and the parallel algorithm, which guarantees the correctness and performance of the proposed algorithm. Our experiments on datasets with O(105) features demonstrate the competitive scalability and the performance compared with the original ITML algorithm.

Online dictionary learning for sparse coding is an effective tool for data analysis. It incrementally learns a set of basis vectors with sparse linear combinations of these vectors when new samples appear. Previous work assumes that the samples embed Gaussian noises, which weaken the power of these methods in handling real applications with non-negative data (e.g., frequency data in word counts). Differently, in this paper, we concentrate on online learning for non-negative dictionary by using moment information for sparse Poisson coding. We exploit the non-negativity of Poisson models to learn a set of non-negative basis vectors and a non-negative sparse linear combination for the moment information of samples. Specifically, we first formulate the online learning problem via the maximum-a-posteriori (MAP) framework. We then propose a novel online algorithm which alternatively updates the sparse-coefficient vector and the basis vectors with non-negativity constraints when a new sample arrives. More importantly, we present sufficient convergence analyses to guarantee the performance of the proposed algorithm, which leads to convergence of a stable dictionary for characterizing the moment information of samples. We finally conduct a series of experiments on word-counts data and image data to show merits of the proposed online algorithm.

Anomaly detection plays an important role in management of modern large-scale distributed systems. Logs, which record system runtime information, are widely used for anomaly detection. Traditionally, developers (or operators) often inspect the logs manually with keyword search and rule matching. The increasing scale and complexity of modern systems, however, make the volume of logs explode, which renders the infeasibility of manual inspection. To reduce manual effort, many anomaly detection methods based on automated log analysis are proposed. However, developers may still have no idea which anomaly detection methods they should adopt, because there is a lack of a review and comparison among these anomaly detection methods. Moreover, even if developers decide to employ an anomaly detection method, re-implementation requires a nontrivial effort. To address these problems, we provide a detailed review and evaluation of six state-of-the-art log-based anomaly detection methods, including three supervised methods and three unsupervised methods, and also release an open-source toolkit allowing ease of reuse. These methods have been evaluated on two publicly-available production log datasets, with a total of 15,923,592 log messages and 365,298 anomaly instances. We believe that our work, with the evaluation results as well as the corresponding findings, can provide guidelines for adoption of these methods and provide references for future development.

Good user interface (UI) design is key to successful mobile apps. UI latency, which can be considered as the time between the commencement of a UI operation and its intended UI update, is a critical consideration for app developers. Current literature still lacks a comprehensive study on how much UI latency a user can tolerate or how to identify UI design defects that cause intolerably long UI latency. As a result, bad UI apps are still common in app markets, leading to extensive user complaints. This paper examines user expectations of UI latency, anddevelops a tool to pinpoint intolerable UI latency in Android apps. To this end, we design an app to conduct a user survey of app UI latency. Through the survey, we find the tendency between user patience and UI latency. Therefore a timely screen update (e.g., loading animations) is critical to heavy-weighted UI operations (i.e. those that incur a long execution time before the final UI update is available). We then design a tool that, by monitoring the UI inputs and updates, can detect apps that do not follow this criterion. The survey and the tool are open-source released on-line. We also apply the tool to many real-world apps. The results demonstrate the effectiveness of the tool in combating app UI design defects.

App developers publish apps on different platforms, such as Google Play, App Store, and Windows Store, to maximize the user volumes and potential revenues. Due to the different characteristics of the platforms and the different user preference (e.g., Android is more customized than iOS), app testing cases on these three platforms should also be designed differently. Comprehensive app testing can be time-consuming for developers. Therefore, understanding the differences of the app issues on these platforms can facilitate the testing process. In this paper, we propose a novel framework named CrossMiner to analyze the essential app issues and explore whether the app issues exhibit differently on the three platforms. Based on five million user reviews, the framework automatically captures the distributions of seven app issues, i.e., "battery", "crash", "memory", "network", "privacy", "spam", and "UI". We discover that the apps for different platforms indeed generate different issue distributions, which can be employed by app developers to schedule and design the testing cases. The verification based on the official user forums also demonstrates the effectiveness of our framework. Furthermore, we also identify that the issues related to "crash" and "network" are more concerned by users than the other issues on these three platforms. To assist developers in gaining a deep insight on the user issues, we also prioritize the user reviews corresponding to the issues. Overall, we aim at understanding the differences of issues on different platforms and facilitating the testing process for app developers.

Rapid UI responsiveness is a key consideration to Android app developers. However, the complicated concurrency model of Android makes it hard for developers to understand and further diagnose the UI performance. This paper presents DiagDroid, a tool specifically designed for Android UI performance diagnosis. The key notion of DiagDroid is that UI-triggered asynchronous executions contribute to the UI performance, and hence their performance and their runtime dependency should be properly captured to facilitate performance diagnosis. However, there are tremendous ways to start asynchronous executions, posing a great challenge to profiling such executions and their runtime dependency. To this end, we properly abstract five categories of asynchronous executions as the building basis. As a result, they can be tracked and profiled based on the specifics of each category with a dynamic instrumentation approach carefully tailored for Android. DiagDroid can then accordingly profile the asynchronous executions in a task granularity, equipping it with low-overhead and high compatibility merits. The tool is successfully applied in diagnosing 33 real-world open-source apps, and we find 14 of them contain 27 performance issues. It shows the effectiveness of our tool in Android UI performance diagnosis. The tool is open-source released online.

An increasing number of companies are beginning to deploy services/applications in the cloud computing environment. Enhancing the reliability of cloud service has become a critical and challenging research problem. In the cloud computing environment, all resources are commercialized. Therefore, a reliability enhancement approach should not consume too much resource. However, existing approaches cannot achieve the optimal effect because of checkpoint image-sharing neglect, and checkpoint image inaccessibility caused by node crashing. To address this problem, we propose a cloud service reliability enhancement approach for minimizing network and storage resource usage in a cloud data center. In our proposed approach, the identical parts of all virtual machines that provide the same service are checkpointed once as the service checkpoint image, which can be shared by those virtual machines to reduce the storage resource consumption. Then, the remaining checkpoint images only save the modified page. To persistently store the checkpoint image, the checkpoint image storage problem is modeled as an optimization problem. Finally, we present an efficient heuristic algorithm to solve the problem. The algorithm exploits the data center network architecture characteristics and the node failure predicator to minimize network resource usage. To verify the effectiveness of the proposed approach, we extend the renowned cloud simulator Cloudsim and conduct experiments on it. Experimental results based on the extended Cloudsim show that the proposed approach not only guarantees cloud service reliability, but also consumes fewer network and storage resources than other approaches.

Location-based social networks (LBSNs), such as Gowalla, Facebook, Foursquare, Brightkite, and so on, have attracted millions of users to share their social friendship and their locations via check-ins in the past few years. Plenty of valuable information is accumulated based on the check-in behaviors, which makes it possible to learn users’ moving patterns as well as their preferences. In LBSNs, point-of-interest (POI) recommendation is one of the most significant tasks because it can help targeted users explore their surroundings as well as help third-party developers provide personalized services. Matrix factorization is a promising method for this task because it can capture users’ preferences to locations and is widely adopted in traditional recommender systems such as movie recommendation. However, the sparsity of the check-in data makes it difficult to capture users’ preferences accurately. Geographical influence can help alleviate this problem and have a large impact on the final recommendation result. By studying users’ moving patterns, we find that users tend to check in around several centers and different users have different numbers of centers. Based on this, we propose a Multi-center Gaussian Model (MGM) to capture this pattern via modeling the probability of a user’s check-in on a location. Moreover, users are usually more interested in the top 20 or even top 10 recommended POIs, which makes personalized ranking important in this task. From previous work, directly optimizing for pairwise ranking like Bayesian Personalized Ranking (BPR) achieves better performance in the top- k recommendation than directly using matrix matrix factorization that aims to minimize the point-wise rating error. To consider users’ preferences, geographical influence and personalized ranking, we propose a unified POI recommendation framework, which unifies all of them together. Specifically, we first fuse MGM with matrix factorization methods and further with BPR using two different approaches. We conduct experiments on Gowalla and Foursquare datasets, which are two large-scale real-world LBSN datasets publicly available online. The results on both datasets show that our unified POI recommendation framework can produce better performance.

With the fast development of Web services in service-oriented systems, the requirement of efficient Quality of Service (QoS) evaluation methods becomes strong. However, many QoS values are unknown in reality. Therefore, it is necessary to predict the unknown QoS values of Web services based on the obtainable QoS values. Generally, the QoS values of similar users are employed to make predictions for the current user. However, the QoS values may be contributed from unreliable users, leading to inaccuracy of the prediction results. To address this problem, we present a highly credible approach, called reputation-based Matrix Factorization (RMF), for predicting the unknown Web service QoS values. RMF first calculates the reputation of each user based on their contributed QoS values to quantify the credibility of users, and then takes the users' reputation into consideration for achieving more accurate QoS prediction. Reputation-based matrix factorization is applicable to the prediction of QoS data in the presence of unreliable user-provided QoS values. Extensive experiments are conducted with real-world Web service QoS data sets, and the experimental results show that our proposed approach outperforms other existing approaches.

Due to the popularity of service-oriented architectures for various distributed systems, an increasing number of Web services have been deployed all over the world. Recently, Web service recommendation became a hot research topic, one that aims to accurately predict the quality of functional satisfactory services for each end user. Generally, the performance of Web service changes over time due to variations of service status and network conditions. Instead of employing the conventional temporal models, we propose a novel spatial-temporal QoS prediction approach for time-aware Web service recommendation, where a sparse representation is employed to model QoS variations. Specifically, we make a zero-mean Laplace prior distribution assumption on the residuals of the QoS prediction, which corresponds to a Lasso regression problem. To effectively select the nearest neighbor for the sparse representation of temporal QoS values, the geo-location of web service is employed to reduce searching range while improving prediction accuracy. The extensive experimental results demonstrate that the proposed approach outperforms state-of-art methods with more than 10% improvement on the accuracy of temporal QoS prediction for time-aware Web service recommendation.

Online learning from imbalanced streaming data to capture the nonlinearity and heterogeneity of the data is significant in machine learning and data mining. To tackle this problem, we propose a kernelized online imbalanced learning (KOIL) algorithm to directly maximize the area under the ROC curve (AUC). We address two more challenges: 1) How to control the number of support vectors without sacrificing model performance; and 2) how to restrict the fluctuation of the learned decision function to attain smooth updating. To this end, we introduce two buffers with fixed budgets (buffer sizes) for positive class and negative class, respectively, to store the learned support vectors, which can allow us to capture the global information of the decision boundary. When determining the weight of a new support vector, we confine its influence only to its $k$-nearest opposite support vectors. This can restrict the effect of new instances and prevent the harm of outliers. More importantly, we design a sophisticated scheme to compensate the model after replacement is conducted when either buffer is full. With this compensation, the learned model approaches the one learned with infinite budgets. We present both theoretical analysis and extensive experimental comparison to demonstrate the effectiveness of our proposed KOIL.

Community detection is an important technique to understand structures and patterns in complex networks. Recently, overlapping community detection becomes a trend due to the ubiquity of overlapping and nested communities in real world. However, existing approaches have ignored the use of implicit link preference information, i.e., links can reflect a node's preference on the targets of connections it wants to build. This information has strong impact on community detection since a node prefers to build links with nodes inside its community than those outside its community. In this paper, we propose a preference-based nonnegative matrix factorization (PNMF) model to incorporate implicit link preference information. Unlike conventional matrix factorization approaches, which simply approximate the original adjacency matrix in value, our model maximizes the likelihood of the preference order for each node by following the intuition that a node prefers its neighbors than other nodes. Our model overcomes the indiscriminate penalty problem in which non-linked pairs inside one community are equally penalized in objective functions as those across two communities. We propose a learning algorithm which can learn a node-community membership matrix via stochastic gradient descent with bootstrap sampling. We evaluate our PNMF model on several real-world networks. Experimental results show that our model outperforms state-of-the-art approaches and can be applied to large datasets.

Logging is a common programming practice of practical importance to collect system runtime information for postmortem analysis. Strategic logging placement is desired to cover necessary runtime information without incurring unintended consequences (e.g., Performance overhead, trivial logs). However, in current practice, there is a lack of rigorous specifications for developers to govern their logging behaviours. Logging has become an important yet tough decision which mostly depends on the domain knowledge of developers. To reduce the effort on making logging decisions, in this paper, we propose a "learning to log" framework, which aims to provide informative guidance on logging during development. As a proof of concept, we provide the design and implementation of a logging suggestion tool, Log Advisor, which automatically learns the common logging practices on where to log from existing logging instances and further leverages them for actionable suggestions to developers. Specifically, we identify the important factors for determining where to log and extract them as structural features, textual features, and syntactic features. Then, by applying machine learning techniques (e.g., Feature selection and classifier learning) and noise handling techniques, we achieve high accuracy of logging suggestions. We evaluate Log Advisor on two industrial software systems from Microsoft and two open-source software systems from Git Hub (totally 19.1M LOC and 100.6K logging statements). The encouraging experimental results, as well as a user study, demonstrate the feasibility and effectiveness of our logging suggestion tool. We believe our work can serve as an important first step towards the goal of "learning to log".

QoS-based Web service recommendation has recently gained much attention for providing a promising way to help users find high-quality services. To facilitate such recommendations, existing studies suggest the use of collaborative filtering techniques for personalized QoS prediction. These approaches, by leveraging partially observed QoS values from users, can achieve high accuracy of QoS predictions on the unobserved ones. However, the requirement to collect users' QoS data likely puts user privacy at risk, thus making them unwilling to contribute their usage data to a Web service recommender system. As a result, privacy becomes a critical challenge in developing practical Web service recommender systems. In this paper, we make the first attempt to cope with the privacy concerns for Web service recommendation. Specifically, we propose a simple yet effective privacy-preserving framework by applying data obfuscation techniques, and further develop two representative privacy-preserving QoS prediction approaches under this framework. Evaluation results from a publicly-available QoS dataset of real-world Web services demonstrate the feasibility and effectiveness of our privacy-preserving QoS prediction approaches. We believe our work can serve as a good starting point to inspire more research efforts on privacy-preserving Web service recommendation.

User review analysis is critical to the bug-fixing and version-modification process for app developers. Many research efforts have been put to user review mining in discovering app issues, including laggy user interface, high memory overhead, privacy leakage, etc. Existing exploration of app reviews generally depends on static collections. As a result, they largely ignore the fact that user reviews are tightly related to app versions. Furthermore, the previous approaches require a developer to spend much time on filtering out trivial comments and digesting the informative textual data. This would be labor-intensive especially to popular apps with tremendous reviews. In the paper, we target at designing a framework in Prioritizing App Issues for Developers (PAID) with minimal manual power and good accuracy. The PAID design is based on the fact that the issues presented in the level of phrase, i.e., a couple of consecutive words, can be more easily understood by developers than in long sentences. Hence, we aim at recommending phrase-level issues of an app to its developers by tracking reviews over the release versions of the app. To assist developers in better comprehending the app issues, PAID employs ThemeRiver to visualize the analytical results to developers. Finally, PAID also allows the developers to check the most related reviews, when they want to obtain a deep insight of a certain issue. In contrast to the traditional evaluation methods such as manual labeling or examining the discussion forum, our experimental study exploits the first-hand information from developers, i.e., app changelogs, to measure the performance of PAID. We analyze millions of user reviews from 18 apps with 117 app versions and the results show that the prioritized issues generated by PAID match the official changelogs with high precision.

A new security problem on smartphones is the wide spread of spyware nested in apps, which occasionally and silently collects user's private data in the background. The state-of-the-art work for privacy leakage detection is dynamic taint analysis, which, however, suffers usability issues because it requires flashing a customized system image to track the taint propagation and consequently incurs great overhead. Through a real-world privacy leakage case study, we observe that the spyware behaviors share some common features during execution, which may further indicate a correlation between the data flow of privacy leakage and some specific features of program execution traces. In this work, we examine such a hypothesis using the newly proposed SpyAware framework, together with a customized TaintDroid as the ground truth. SpyAware includes a profiler to automatically profile app executions in binder calls and system calls, a feature extractor to extract feature vectors from execution traces, and a classifier to train and predict spyware executions based on the feature vectors. We conduct an evaluation experiment with 100 popular apps downloaded from Google Play. Experimental results show that our approach can achieve promising performance with 67.4% accuracy in detecting device id spyware executions and 78.4% in recognizing location spyware executions.

The SDA workshop at WSDM 2015 is the fifth International Workshop on Scalable Data Analytics, following the previous four workshops of SDA respectively held at IEEE Big Data 2013, PAKDD 2014, IEEE Big Data 2014, and IEEE ICDM 2014. This series of workshops aims to provide professionals, researchers, and technologists with a single forum where they can discuss and share the state-of-the-art theories and applications of scalable data analytics technologies. In particular, in the era of information explosion, the scientific, biomedical, and engineering research communities are undergoing a profound transformation where discoveries and innovations increasingly rely on massive amounts of data. The characteristics of volume, velocity, variety and veracity originated in the massive big data then bring challenges to current data analytics techniques. The focus of the fifth SDA is to discuss how we can scale up data analytics techniques for modeling and analyzing big data from various domains.

Functionally equivalent web services can be composed to form more reliable service-oriented systems. However, the choice of fault tolerance strategy can have a significant effect on the quality-of-service (QoS) of the resulting service-oriented systems. In this paper, we investigate the problem of selecting an optimal fault tolerance strategy for building reliable service-oriented systems. We formulate the user requirements as local and global constraints and model the selection of fault tolerance strategy as an optimization problem. A heuristic algorithm is proposed to efficiently solve the optimization problem. Fault tolerance strategy selection for semantically related tasks is also investigated in this paper. Large-scale real-world experiments are conducted to illustrate the benefits of the proposed approach. The experimental results show that our problem modeling approach and the proposed selection algorithm make it feasible to manage the fault tolerance of complex service-oriented systems both efficiently and effectively.

This paper presents compact yet comprehensive feature representations for the electroencephalogram (EEG) signal to achieve efficient epileptic seizure prediction performance. The initial EEG feature vectors are formed by acquiring the dominant amplitude and frequency components on an epoch-by-epoch basis from the EEG signals. These extracted parameters can reveal the intrinsic EEG signal changes as well as the underlying stage transitions. To improve the efficacy of feature extraction, an elimination-based feature selection method has been applied on the initial feature vectors. This diminishes redundant and noisy points, providing each patient with a lower dimensional and independent final feature form. In this context, our study is distinguished from that of others currently prevailing. Usually, these latter approaches adopted feature extraction processes, which employed time-consuming high-dimensional parameter sets. Machine learning approaches that are considered as state of the art have been employed to build patient-specific binary classifiers that can divide the extracted feature parameters into preictal and interictal groups. Through out-of-sample evaluation on the intracranial EEG recordings provided by the publicly available Freiburg dataset, promising prediction performance has been attained. Specifically, we have achieved 98.8% sensitivity results on the 19 patients included in our experiment, where only one of 83 seizures across all patients was not predicted. To make this investigation more comprehensive, we have conducted extensive comparative studies with other recently published competing approaches, in which the advantages of our method are highlighted.

Recommender systems are promising for providing personalized favorite services. Collaborative filtering (CF) technologies, making prediction of users' preference based on users' previous behaviors, have become one of the most successful techniques to build modern recommender systems. Several challenging issues occur in previously proposed CF methods: (1) most CF methods ignore users' response patterns and may yield biased parameter estimation and suboptimal performance; (2) some CF methods adopt heuristic weight settings, which lacks a systematical implementation; and (3) the multinomial mixture models may weaken the computational ability of matrix factorization for generating the data matrix, thus increasing the computational cost of training. To resolve these issues, we incorporate users' response models into the probabilistic matrix factorization (PMF), a popular matrix factorization CF model, to establish the response aware probabilistic matrix factorization (RAPMF) framework. More specifically, we make the assumption on the user response as a Bernoulli distribution which is parameterized by the rating scores for the observed ratings while as a step function for the unobserved ratings. Moreover, we speed up the algorithm by a mini-batch implementation and a crafting scheduling policy. Finally, we design different experimental protocols and conduct systematical empirical evaluation on both synthetic and real-world datasets to demonstrate the merits of the proposed RAPMF and its mini-batch implementation.

Recently, hybrid cloud computing paradigm has be widely advocated as a promising solution for Software-as-a-Service (SaaS) providers to effectively handle the dynamic user requests. With such a paradigm, the SaaS providers can extend their local services into the public clouds seamlessly so that the dynamic user request workload to a SaaS can be elegantly processed with both the local servers and the rented computing capacity in the public cloud. However, although it is suggested that a hybrid cloud may save cost compared with building a powerful private cloud, considerable renting cost and communication cost are still introduced in such a paradigm. How to optimize such operational cost becomes one major concern for the SaaS providers to adopt the hybrid cloud computing paradigm. However, this critical problem remains unanswered in the current state of the art. In this paper, we focus on optimizing the operational cost for the hybrid cloud paradigm by theoretically analyzing the problem with a Lyapunov optimization framework. This allows us to design an online dynamic provision algorithm. In this way, our approach can address the real-world challenges where no a priori information of public cloud renting prices is available and the future probability distribution of user requests is unknown. We then conduct extensive experimental study based on a set of real-world data, and the results confirm that our algorithm can work effectively in reducing the operational cost.

As the number of Web services surges rapidly, recommending the right service for various users on demand has become one of the most challenging research issues in the fields of services computing and cloud computing. This special issue focuses on the recommendation techniques for services computing and cloud computing including classic recommendation algorithms for services computing and cloud computing, emerging recommendation techniques for service selection and composition, and applications of recommendation techniques in composition of complex service mashups, ad-hoc social networks, and green cloud environment. The papers introduced in this special issue illustrate the efficiency and effectiveness of services computing and cloud computing, while demonstrating a variety of challenges that arise. It is expected that this special issue, as a whole, will provide integrated and synthesized view of the current state of the art, identify key challenges, possible research directions and opportunities for investigation, and promote community-building efforts among researchers and practitioners in the related fields.

Web service recommendation systems can help service users to locate the right service from the large number of available web services. Avoiding recommending dishonest or unsatisfactory services is a fundamental research problem in the design of web service recommendation systems. Reputation of web services is a widely-employed metric that determines whether the service should be recommended to a user. The service reputation score is usually calculated using feedback ratings provided by users. Although the reputation measurement of web service has been studied in the recent literature, existing malicious and subjective user feedback ratings often lead to a bias that degrades the performance of the service recommendation system. In this paper, we propose a novel reputation measurement approach for web service recommendations. We first detect malicious feedback ratings by adopting the cumulative sum control chart, and then we reduce the effect of subjective user feedback preferences employing the Pearson Correlation Coefficient. Moreover, in order to defend malicious feedback ratings, we propose a malicious feedback rating prevention scheme employing Bloom filtering to enhance the recommendation performance. Extensive experiments are conducted by employing a real feedback rating data set with 1.5 million web service invocation records. The experimental results show that our proposed measurement approach can reduce the deviation of the reputation measurement and enhance the success ratio of the web service recommendation.

Trace-oriented monitoring is one of the main methods for monitoring cloud systems. However, there is no free trace data set available, which hinders the development of trace-oriented monitoring. Therefore, we want to collect a trace data set in a real environment and make it free. During collection, many aspects are considered, including cluster sizes, user requests, workload speeds, injected faults, etc., to simulate different situations. The structure of this data set is well-designed. We believe that this data set will be helpful for the research of trace-oriented monitoring.

User request trace-oriented monitoring is an effective method to improve the reliability of cloud systems. However, there are some difficulties in getting traces in practice, which hinder the development of trace-oriented monitoring research. In this paper, we release a fine-grained user request-centric open trace data set, called Trace Bench, collected on a real world cloud storage system deployed in a real environment. During collecting, many aspects are considered to simulate different scenarios, including cluster size, request type, workload speed, etc. Besides recording the traces when the monitored system is running normally, we also collect the traces under the situation with faults injected. With a mature injection tool, 14 faults are introduced, including function faults and performance faults. The traces in Trace Bench are clustered in different files, where each file corresponds to a certain scenario. The whole collection work lasted for more than half a year, resulting in more than 360, 000 traces in 361 files. In addition, we also employ several applications based on Trace Bench, which validate the helpfulness of Trace Bench for the field of trace-oriented monitoring.

Service-based cloud applications are typically built on component services to fulfill certain application logic. To meet quality-of-service (QoS) guarantees, these applications have to become resilient against the QoS variations of their component services. Runtime service adaptation has been recognized as a key solution to achieve this goal. To make timely and accurate adaptation decisions, effective QoS prediction is desired to obtain the QoS values of component services. However, current research has focused mostly on QoS prediction of the working services that are being used by a cloud application, but little on QoS prediction of candidate services that are also important for making adaptation decisions. To bridge this gap, in this paper, we propose a novel QoS prediction approach, namely adaptive matrix factorization (AMF), which is inspired from the collaborative filtering model used in recommender systems. Specifically, our AMF approach extends conventional matrix factorization into an online, accurate, and scalable model by employing techniques of data transformation, online learning, and adaptive weights. Comprehensive experiments have been conducted based on a real-world large-scale QoS dataset of Web services to evaluate our approach. The evaluation results provide good demonstration for our approach in achieving accuracy, efficiency, and scalability.

Web service recommendation is of great importance when users face a large number of functionally-equivalent candidate services. To recommend Web services that best fit a user's need, QoS values which characterize the non-functional properties of those candidate services are in demand. But in reality, the QoS information of Web service is not easy to obtain, because only limited historical invocation records exist. To tackle this challenge, in recent literature, a number of QoS prediction methods are proposed, but they still demonstrate disadvantages on prediction accuracy. In this paper, we design a location-based hierarchical matrix factorization (HMF) method to perform personalized QoS prediction, whereby effective service recommendation can be made. We cluster users and services into several user-service groups based on their location information, each of which contains a small set of users and services. To better characterize the QoS data, our HMF model is trained in a hierarchical way by using the global QoS matrix as well as several location-based local QoS matrices generated from user-service clusters. Then the missing QoS values can be predicted by compactly combining the results from local matrix factorization and global matrix factorization. Comprehensive experiments are conducted on a real-world Web service QoS dataset with 1,974,675 real Web service invocation records. The experimental results show that our HMF method achieves higher prediction accuracy than the state-of-the-art methods.

Hybrid cloud computing paradigm has recently be widely advocated, where Software-as-a-Service (SaaS) providers can extend their local services into the public clouds seamlessly. In this way, dynamic user request workload to a SaaS can be elegantly handled with the rented computing capacity in public cloud. However, although a hybrid cloud may save cost compared with the private cloud, it still introduces considerable renting cost and communication cost. How to optimize such an operational cost becomes one major concern for the SaaS providers to adopt such a hybrid cloud computing paradigm. However, this critical problem remains unanswered in the current state of the art. In this paper, we focus on optimizing the operational cost for the hybrid cloud model by theoretically analyzing the problem with a Lyapunov optimization framework, and accordingly providing an online dynamic provision algorithm. In this way, our approach can address the real-world challenges where no a priori information of public cloud renting prices is available and the future probability distribution of user requests is unknown. We then conduct experimental study based on a set of real-world data, and the results confirm that our algorithm can work well in reducing the cost.

Social information between users has been widely used to improve the traditional Recommender System in many previous works. However, in many websites such as Amazon and eBay, there is no explicit social graph that can be used to improve the recommendation performance. Hence in this work, in order to make it possible to employ social recommendation methods in those non-social information websites, we propose a general framework to construct a homophily-based implicit social network by utilizing both the rating and comments of items given by the users. Our scalable framework can be easily extended to enhance the performance of any recommender systems without social network by replacing the homophily-based implicit social relation definition. We propose four methods to extract and analyze the implicit social links between users, and then conduct the experiments on Amazon dataset. Experimental results show that our proposed methods work better than traditional recommendation methods without social information.

Recommendation techniques have been well developed in the past decades. Most of them build models only based on user item rating matrix. However, in real world, there is plenty of auxiliary information available in recommendation systems. We can utilize these information as additional features to improve recommendation performance. We refer to recommendation with auxiliary information as context-aware recommendation. Context-aware Factorization Machines (FM) is one of the most successful context-aware recommendation models. FM models pairwise interactions between all features, in such way, a certain feature latent vector is shared to compute the factorized parameters it involved. In practice, there are tens of context features and not all the pairwise feature interactions are useful. Thus, one important challenge for context-aware recommendation is how to effectively select "good" interaction features. In this paper, we focus on solving this problem and propose a greedy interaction feature selection algorithm based on gradient boosting. Then we propose a novel Gradient Boosting Factorization Machine (GBFM) model to incorporate feature selection algorithm with Factorization Machines into a unified framework. The experimental results on both synthetic and real datasets demonstrate the efficiency and effectiveness of our algorithm compared to other state-of-the-art methods.

Most existing recommender systems focus on modeling the ratings while ignoring the abundant information embedded in the review text. In this paper, we propose a unified model that combines content-based filtering with collaborative filtering, harnessing the information of both ratings and reviews. We apply topic modeling techniques on the review text and align the topics with rating dimensions to improve prediction accuracy. With the information embedded in the review text, we can alleviate the cold-start problem. Furthermore, our model is able to learn latent topics that are interpretable. With these interpretable topics, we can explore the prior knowledge on items or users and recommend completely "cold"' items. Empirical study on 27 classes of real-life datasets show that our proposed model lead to significant improvement compared with strong baseline methods, especially for datasets which are extremely sparse where rating-only methods cannot make accurate predictions.

Web services are integrated software components for the support of interoperable machine-to-machine interaction over a network. Web services have been widely employed for building service-oriented applications in both industry and academia in recent years. The number of publicly available Web services is steadily increasing on the Internet. However, this proliferation makes it hard for a user to select a proper Web service among a large amount of service candidates. An inappropriate service selection may cause many problems (e.g., ill-suited performance) to the resulting applications. In this paper, we propose a novel collaborative filtering-based Web service recommender system to help users select services with optimal Quality-of-Service (QoS) performance. Our recommender system employs the location information and QoS values to cluster users and services, and makes personalized service recommendation for users based on the clustering results. Compared with existing service recommendation methods, our approach achieves considerable improvement on the recommendation accuracy. Comprehensive experiments are conducted involving more than 1.5 million QoS records of real-world Web services to demonstrate the effectiveness of our approach.

Software systems that run continuously over a long time have been frequently reported encountering gradual degradation issues. That is, as time progresses, software tends to exhibit degraded performance, deflated service capacity, or deteriorated QoS. Currently, the state-of-the-art approach of Mann-Kendall Test & Seasonal Kendall Test & Sen's Slope Estimator & Seasonal Sen's Slope Estimator (MKSK) detects and characterizes degradation via a combination of techniques in statistical trend analysis. Nevertheless, we pinpoint some drawbacks of MKSK in this paper: 1) MKSK cannot be automated for large scale software degradation analysis, 2) MKSK estimates the degradation trend of software in an oversimplified linear way, 3) MKSK is sensitive to noise, and 4) MKSK suffers from high computational complexity. To overcome all these limitations, we propose a more advanced approach called Modified Cox-Stuart Test & Iterative Hodrick-Prescott Filter (CSHP). The superiority of our CSHP approach over MKSK is validated through extensive Monte Carlo simulations, as well as a real performance dataset measured from 99 real-world web servers.

The on-demand use, high scalability, and low maintenance cost nature of cloud computing have attracted more and more enterprises to migrate their legacy applications to the cloud environment. Although the cloud platform itself promises high reliability, ensuring high quality of service is still one of the major concerns, since the enterprise applications are usually complicated and consist of a large number of distributed components. Thus, improving the reliability of an application during cloud migration is a challenging and critical research problem. To address this problem, we propose a reliability-based optimization framework, named ROCloud, to improve the application reliability by fault tolerance. ROCloud includes two ranking algorithms. The first algorithm ranks components for the applications that all their components will be migrated to the cloud. The second algorithm ranks components for hybrid applications that only part of their components are migrated to the cloud. Both algorithms employ the application structure information as well as the historical reliability information for component ranking. Based on the ranking result, optimal fault-tolerant strategy will be selected automatically for the most significant components with respect to their predefined constraints. The experimental results show that by refactoring a small number of error-prone components and tolerating faults of the most significant components, the reliability of the application can be greatly improved.

Quality of service (QoS) is widely employed for describing nonfunctional characteristics of web services. Although QoS of web services has been investigated intensively in the field of service computing, there is a lack of real-world web service QoS data sets for validating various QoS-based techniques and models. To investigate QoS of real-world web services and to provide reusable research data sets for future research, we conduct several large-scale evaluations on real-world web services. First, addresses of 21,358 web services are obtained from the Internet. Then, three large-scale real-world evaluations are conducted. In our evaluations, more than 30 million real-world web service invocations are conducted on web services in more than 80 countries by users from more than 30 counties. Detailed evaluation results are presented in this paper and comprehensive web service QoS data sets are publicly released online.

The exponential growth of Web service makes building high-quality service-oriented systems an urgent and crucial research problem. Performance of the service-oriented systems highly depends on the remote Web services as well as the unpredictability of the Internet. Performance prediction of service-oriented systems is critical for automatically selecting the optimal Web service composition. Since the performance of Web services is highly related to the service status and network environments which are variable over time, it is an important task to predict the performance of service-oriented systems at run-time. To address this critical challenge, this paper proposes an online performance prediction framework, called OPred, to provide personalized service-oriented system performance prediction efficiently. Based on the past usage experience from different users, OPred builds feature models and employs time series analysis techniques on feature trends to make performance prediction. The results of large-scale real-world experiments show the effectiveness and efficiency of OPred.

Application latency is one significant user metric for evaluating the performance of online cloud applications. However, as applications are migrated to the cloud and deployed across a wide-area network, the application latency usually presents high variability over time. Among lots of subtleties that influence the latency, one important factor is relying on the Internet for application connectivity, which introduces a high degree of variability and uncertainty on user-perceived application latency. As a result, a key challenge faced by application designers is how to build consistently low-latency cloud applications with the large number of geo-distributed and latency-varying cloud components. In this paper, we propose a dynamic request routing framework, DR2, by taking full advantage of redundant components in the clouds to tolerate latency variability. In practice, many functionally-equivalent components have been already deployed redundantly for load balancing and fault tolerance, thus resulting in low additional overhead for DR2. To evaluate the performance of our approach, we conduct a set of experiments based on two large-scale real-world datasets and a synthetic dataset. The results show the effectiveness and efficiency of our approach.

Document clustering plays an important role in large scale textual data analysis, which generally faces with great challenge of the high dimensional textual data. One remedy is to learn the high-level sparse representation by the sparse coding techniques. In contrast to traditional Gaussian noise-based sparse coding methods, in this paper, we employ a Poisson distribution model to represent the word-count frequency feature of a text for sparse coding. Moreover, a novel sparse-constrained Poisson regression algorithm is proposed to solve the induced optimization problem. Different from previous Poisson regression with the family of ℓ1-regularization to enhance the sparse solution, we introduce a sparsity ratio measure which make use of both ℓ1-norm and ℓ2-norm on the learned weight. An important advantage of the sparsity ratio is that it bounded in the range of 0 and 1. This makes it easy to set for practical applications. To further make the algorithm trackable for the high dimensional textual data, a projected gradient descent algorithm is proposed to solve the regression problem. Extensive experiments have been conducted to show that our proposed approach can achieve effective representation for document clustering compared with state-of-the-art regression methods.

Social media like forums and microblogs have accumulated a huge amount of user generated content (UGC) containing human knowledge.Currently, most of UGC is listed as a whole or in pre-defined categories.This "list-based" approach is simple, but hinders users from browsing and learning knowledge of certain topics effectively.To address this problem, we propose a hierarchical entity-based approach for structuralizing UGC in social media.By using a large-scale entity repository, we design a three-step framework to organize UGC in a novel hierarchical structure called "cluster entity tree (CET)".With Yahoo!Answers as a test case, we conduct experiments and the results show the effectiveness of our framework in constructing CET.We further evaluate the performance of CET on UGC organization in both user and system aspects.From a user aspect, our user study demonstrates that, with CET-based structure, users perform significantly better in knowledge learning than using traditional list-based approach.From a system aspect, CET substantially boosts the performance of two information retrieval models (i.e., vector space model and query likelihood language model).

Tensor completion from incomplete observations is a problem of significant practical interest. However, it is unlikely that there exists an efficient algorithm with provable guarantee to recover a general tensor from a limited number of observations. In this paper, we study the recovery algorithm for pairwise interaction tensors, which has recently gained considerable attention for modeling multiple attribute data due to its simplicity and effectiveness. Specifically, in the absence of noise, we show that one can exactly recover a pairwise interaction tensor by solving a constrained convex program which minimizes the weighted sum of nuclear norms of matrices from $O(nr\log^2(n))$ observations. For the noisy cases, we also prove error bounds for a constrained convex program for recovering the tensors. Our experiments on the synthetic dataset demonstrate that the recovery performance of our algorithm agrees well with the theory. In addition, we apply our algorithm on a temporal collaborative filtering task and obtain state-of-the-art results.

The context-free language (CFL) reachability problem is a well-known fundamental formulation in program analysis. In practice, many program analyses, especially pointer analyses, adopt a restricted version of CFL-reachability, Dyck-CFL-reachability, and compute on edge-labeled bidirected graphs. Solving the all-pairs Dyck-CFL-reachability on such bidirected graphs is expensive. For a bidirected graph with n nodes and m edges, the traditional dynamic programming style algorithm exhibits a subcubic time complexity for the Dyck language with k kinds of parentheses. When the underlying graphs are restricted to bidirected trees, an algorithm with O(n log n log k) time complexity was proposed recently. This paper studies the Dyck-CFL-reachability problems on bidirected trees and graphs. In particular, it presents two fast algorithms with O(n) and O(n + m log m) time complexities on trees and graphs respectively. We have implemented and evaluated our algorithms on a state-of-the-art alias analysis for Java. Results on standard benchmarks show that our algorithms achieve orders of magnitude speedup and consume less memory.

Learning explanatory features across multiple related tasks, or MultiTask Feature Selection (MTFS), is an important problem in the applications of data mining, machine learning, and bioinformatics. Previous MTFS methods fulfill this task by batch-mode training. This makes them inefficient when data come sequentially or when the number of training data is so large that they cannot be loaded into the memory simultaneously. In order to tackle these problems, we propose a novel online learning framework to solve the MTFS problem. A main advantage of the online algorithm is its efficiency in both time complexity and memory cost. The weights of the MTFS models at each iteration can be updated by closed-form solutions based on the average of previous subgradients. This yields the worst-case bounds of the time complexity and memory cost at each iteration, both in the order ofO(d×Q), wheredis the number of feature dimensions andQis the number of tasks. Moreover, we provide theoretical analysis for the average regret of the online learning algorithms, which also guarantees the convergence rate of the algorithms. Finally, we conduct detailed experiments to show the characteristics and merits of the online learning algorithms in solving several MTFS problems.

Service Oriented Architecture (SOA) is a business-centric IT architectural approach for building distributed systems. Reliability of service-oriented systems heavily depends on the remote Web services as well as the unpredictable Internet connections. Designing efficient and effective reliability prediction approaches of Web services has become an important research issue. In this article, we propose two personalized reliability prediction approaches of Web services, that is, neighborhood-based approach and model-based approach. The neighborhood-based approach employs past failure data of similar neighbors (either service users or Web services) to predict the Web service reliability. On the other hand, the model-based approach fits a factor model based on the available Web service failure data and use this factor model to make further reliability prediction. Extensive experiments are conducted with our real-world Web service datasets, which include about 23 millions invocation results on more than 3,000 real-world Web services. The experimental results show that our proposed reliability prediction approaches obtain better reliability prediction accuracy than other competing approaches.

Performance diagnosis is labor intensive in production cloud computing systems. Such systems typically face many real-world challenges, which the existing diagnosis techniques for such distributed systems cannot effectively solve. An efficient, unsupervised diagnosis tool for locating fine-grained performance anomalies is still lacking in production cloud computing systems. This paper proposes CloudDiag to bridge this gap. Combining a statistical technique and a fast matrix recovery algorithm, CloudDiag can efficiently pinpoint fine-grained causes of the performance problems, which does not require any domain-specific knowledge to the target system. CloudDiag has been applied in a practical production cloud computing systems to diagnose performance problems. We demonstrate the effectiveness of CloudDiag in three real-world case studies.

Cloud computing is becoming popular. Building high-quality cloud applications is a critical research problem. QoS rankings provide valuable information for making optimal cloud service selection from a set of functionally equivalent service candidates. To obtain QoS values, real-world invocations on the service candidates are usually required. To avoid the time-consuming and expensive real-world service invocations, this paper proposes a QoS ranking prediction framework for cloud services by taking advantage of the past service usage experiences of other consumers. Our proposed framework requires no additional invocations of cloud services when making QoS ranking prediction. Two personalized QoS ranking prediction approaches are proposed to predict the QoS rankings directly. Comprehensive experiments are conducted employing real-world QoS data, including 300 distributed users and 500 real-world web services all over the world. The experimental results show that our approaches outperform other competing approaches.

With the increasing amount of web services on the Internet, personalized web service selection and recommendation are becoming more and more important. In this paper, we present a new similarity measure for web service similarity computation and propose a novel collaborative filtering approach, called normal recovery collaborative filtering, for personalized web service recommendation. To evaluate the web service recommendation performance of our approach, we conduct large-scale real-world experiments, involving 5,825 real-world web services in 73 countries and 339 service users in 30 countries. To the best of our knowledge, our experiment is the largest scale experiment in the field of service computing, improving over the previous record by a factor of 100. The experimental results show that our approach achieves better accuracy than other competing approaches.

With the increasing presence and adoption of web services on the World Wide Web, the demand of efficient web service quality evaluation approaches is becoming unprecedentedly strong. To avoid the expensive and time-consuming web service invocations, this paper proposes a collaborative quality-of-service (QoS) prediction approach for web services by taking advantages of the past web service usage experiences of service users. We first apply the concept of user-collaboration for the web service QoS information sharing. Then, based on the collected QoS data, a neighborhood-integrated approach is designed for personalized web service QoS value prediction. To validate our approach, large-scale real-world experiments are conducted, which include 1,974,675 web service invocations from 339 service users on 5,825 real-world web services. The comprehensive experimental studies show that our proposed approach achieves higher prediction accuracy than other approaches. The public release of our web service QoS data set provides valuable real-world data for future research.

Nowadays, more and more scientific applications are moving to cloud computing. The optimal deployment of scientific applications is critical for providing good services to users. Scientific applications are usually topology-aware applications. Therefore, considering the topology of a scientific application during the development will benefit the performance of the application. However, it is challenging to automatically discover and make use of the communication pattern of a scientific application while deploying the application on cloud. To attack this challenge, in this paper, we propose a framework to discover the communication topology of a scientific application by pre-execution and multi-scale graph clustering, based on which the deployment can be optimized. Comprehensive experiments are conducted by employing a well-known MPI benchmark and comparing the performance of our method with those of other methods. The experimental results show the effectiveness of our topology-aware deployment method.

Cloud computing attracts considerable attention from both industry and academic these years. Nowadays, a number of research investigations have been conducted on cloud-based services (e.g., IaaS, PaaS, SaaS, etc.). Deployment of cloud-based services is one of the most important research problems. In cloud computing, multiple services tend to cooperate with each other to accomplish complicated tasks. Deploying these services independently may not lead to good overall performance, since there are a lot of interactions among different services. Making an optimal co-deployment of multiple services is critical for reducing latency of user requests. When deploying highly related services, taking only distribution of users into consideration is not enough, since the deployment of one service would affect others. To attack this challenge, we employ cross service information as well as user locations to build a new model in integer programming formulation. To reduce the computation time of the model, we purpose a sequential model running iteratively to obtain approximate solution. Extensive experiments have been conducted over a large real-world dataset, involving 307 distributed computers in about 40 countries, and 1881 real-world Internet-based services in about 60 countries. The experimental results show the effectiveness of our proposed model. Our real-world dataset is publicly released to promote future research, which also makes our experiments reproducible.

Recently, location-based social networks (LBSNs), such as Gowalla, Foursquare, Facebook, and Brightkite, etc., have attracted millions of users to share their social friendship and their locations via check-ins. The available check-in information makes it possible to mine users’ preference on locations and to provide favorite recommendations. Personalized Point-of-interest (POI) recommendation is a significant task in LBSNs since it can help targeted users explore their surroundings as well as help third-party developers to provide personalized services. To solve this task, matrix factorization is a promising tool due to its success in recommender systems. However, previously proposed matrix factorization (MF) methods do not explore geographical influence, e.g., multi-center check-in property, which yields suboptimal solutions for the recommendation. In this paper, to the best of our knowledge, we are the first to fuse MF with geographical and social influence for POI recommendation in LBSNs. We first capture the geographical influence via modeling the probability of a user’s check-in on a location as a Multi-center Gaussian Model (MGM). Next, we include social information and fuse the geographical influence into a generalized matrix factorization framework. Our solution to POI recommendation is efficient and scales linearly with the number of observations. Finally, we conduct thorough experiments on a large-scale real-world LBSNs dataset and demonstrate that the fused matrix factorization framework with MGM utilizes the distance information sufficiently and outperforms other state-of-the-art methods significantly.

Automatic Subjective Question Answering (ASQA), which aims at answering users'subjective questions using summaries of multiple opinions, becomes increasingly important. One challenge of ASQA is that expected answers for subjective questions may not readily exist in the Web. The rising and popularity of Community Question Answering (CQA) sites, which provide platforms for people to post and answer questions, provides an alternative to ASQA. One important task of ASQA is question subjectivity identification, which identifies whether a user is asking a subjective question. Unfortunately, there has been little labeled training data available for this task. In this paper, we propose an approach to collect training data automatically by utilizing social signals in CQA sites without involving any manual labeling. Experimental results show that our data-driven approach achieves 9.37% relative improvement over the supervised approach using manually labeled data, and achieves 5.15% relative gain over a state-of-the-art semi-supervised approach. In addition, we propose several heuristic features for question subjectivity identification. By adding these features, we achieve 11.23% relative improvement over word n-gram feature under the same experimental setting.

For the purpose of effective epileptic seizure prediction, this paper presents a new representation for the electroencephalogram (EEG) signal by recurring to their primary amplitude and frequency components. This formulation is then applied on an epoch-by-epoch basis to the preictal and interictal states of EEG signals in order to extract the most dominant amplitude and frequency characteristics. Through inspecting and identifying the distinctive EEG signal changes and stage transitions revealed by these extracted feature vectors, upcoming epileptic seizures could be predicted in due course. Machine learning approaches have been employed to construct patient-specific classifiers that can divide the extracted feature vectors into preictal and interictal groups. In this context, our work is distinguished from most currently adopted feature extraction process which employs time-consuming high-dimensional parameter sets. We have made special efforts to derive discriminative and comprehensive features for the front-end of the epileptic seizure prediction algorithm. To reduce false alarms due to trivial signal fluctuation, a simple yet effective post-processing step is incorporated thereafter. Performance of the prediction algorithm is assessed through out-of-sample evaluation on the intracranial EEG (iEEG) recordings provided by the publicly available Freiburg data set. It has been shown by simulation results that the proposed feature estimation method leads to promising prediction results in terms of sensitivity and specificity. In particular, only four out of the 83 seizures across all the patients included in our experiment were missed by the prediction, which means that a sensitivity as high as 95.2% has been achieved.

The polarity of opinion is a crucial part of information and ignoring the asymmetry between them, can potentially result in an inaccurate estimation of the number of product adoptions and incorrect recommendations. We analyze the propagation patterns of the negative and positive opinions on two real world datasets, Flixster and Epinions, and observe that the presence of negative opinions significantly reduces the number of expressed opinions. To account for the asymmetry between the two kind of opinions, we propose extensions of the two most popular information propagation models, Independent Cascade and Linear Threshold models. The proposed extensions give a tractable influence problem and improves the prediction accuracy of future opinions, by more than 3% on Flixster and 5% on Epinions datasets.

Traditionally click models predict click-through rate (CTR) of an advertisement (ad) independent of other ads. Recent researches however indicate that the CTR of an ad is dependent on the quality of the ad itself but also of the neighboring ads. Using historical click-through data of a commercially available ad server, we identify two types (competing and collaborating) of influences among sponsored ads and further propose a novel click-model, Full Relation Model (FRM), which explicitly models dependencies between ads. On a test data, FRM shows significant improvement in CTR prediction as compared to earlier click models.

In large-scale cloud computing systems, the growing scale and complexity of component interactions pose great challenges for operators to understand the characteristics of system performance. Performance profiling has long been proved to be an effective approach to performance analysis; however, existing approaches do not consider two new requirements that emerge in cloud computing systems. First, the efficiency of the profiling becomes of critical concern; second, visual analytics should be utilized to make profiling results more readable. To address the above two issues, in this paper, we present P-Tracer, an online performance profiling approach specifically tailored for large-scale cloud computing systems. P-Tracer constructs a specific search engine that adopts a proactive way to process performance logs and generates particular indices for fast queries; furthermore, PTracer provides users with a suite of web-based interfaces to query statistical information of all kinds of services, which helps them quickly and intuitively understand system behavior. The approach has been successfully applied in Alibaba Cloud Computing Inc. to conduct online performance profiling both in production clusters and test clusters. Experience with one real-world case demonstrates that P-Tracer can effectively and efficiently help users conduct performance profiling and localize the primary causes of performance anomalies.

With the rapid growth of Web services in recent years, the optimal service selection from functionally-equivalent service candidates has become more critical for building high quality service-oriented systems. To provide accurate QoSvalues for service selection, user-side QoS prediction thus becomes an important research problem. Although collaborative filtering based prediction approaches have been studied in several previous works, these methods suffer from the limitation of the sparsity of available historical QoS data, which greatly degrades the prediction accuracy. To address this problem, this paper proposes a Web service positioning (WSP)framework for response time prediction, which is one of the most important QoS properties. In our approach, a small set of landmarks are deployed to periodically monitor the response times of the Web service candidates and provide references to the numerous service users. By combining the advantages of network coordinate based approaches and collaborative filtering based approaches, the response times between users and Web services can be accurately predicted using their corresponding Euclidean distances. Extensive experiments are conducted based on our real-world QoS dataset collected on PlanetLab, comprising about 359,400 response time values from 200 users on 1,597 Web services. The experimental results show that our WSP approach outperforms the other existing approaches, especially when the historical data is sparse.

Community Question Answering (CQA) attracts increasing volume of research on question retrieval, high quality content discovery and experts finding. However, few studies are focused on community per se of CQA services and also provide an in-depth analysis of them. This paper aims to enrich our knowledge on two of these CQA services, namely Yahoo! Answers and Baidu Zhidao through reviewing their communities, comparing similarities and differences of the two communities, together with analyzing their influence on solving questions. Six data sets are employed for comparative analysis. In this paper: (1) We analyze the social network structures of Yahoo! Answers and Baidu Zhidao; (2) We compare the the social community characteristics of top contributors; (3) We reveal the behaviors of users in different categories in these two portals; (4) We reveal temporal trends of these characteristics; (5) We find that the community of Yahoo! Answers and Baidu Zhidao complement each other in efficiency and effectiveness of answering questions.

Collaborative filtering (CF), aiming at predicting users' unknown preferences based on observational preferences from some users, has become one of the most successful methods to building recommender systems. Various approaches to CF have been proposed in this area, but seldom do they consider the dynamic scenarios: 1) new items arriving in the system, 2) new users joining the system; or 3) new rating updating the system are all dynamically obtained with respect to time. To capture these changes, in this paper, we develop an online learning framework for collaborative filtering. Specifically, we construct this framework consisting of two state-of-the-art matrix factorization based CF methods: the probabilistic matrix factorization and the top-one probability based ranking matrix factorization. Moreover, we demonstrate that the proposed online algorithms bring several attractive advantages: 1) they scale linearly with the number of observed ratings and the size of latent features; 2) they obviate the need to load all ratings in memory; 3) they can adapt to new ratings easily. Finally, we conduct a series of detailed experiments on real-world datasets to demonstrate the merits of the proposed online learning algorithms under various settings.

As the exponential explosion of various contents generated on the Web, Recommendation techniques have become increasingly indispensable. Innumerable different kinds of recommendations are made on the Web every day, including movies, music, images, books recommendations, query suggestions, tags recommendations, etc. No matter what types of data sources are used for the recommendations, essentially these data sources can be modeled in the form of various types of graphs. In this paper, aiming at providing a general framework on mining Web graphs for recommendations, (1) we first propose a novel diffusion method which propagates similarities between different nodes and generates recommendations; (2) then we illustrate how to generalize different recommendation problems into our graph diffusion framework. The proposed framework can be utilized in many recommendation tasks on the World Wide Web, including query suggestions, tag recommendations, expert finding, image recommendations, image annotations, etc. The experimental analysis on large data sets shows the promising future of our work.

Cloud computing is becoming a mainstream aspect of information technology. More and more enterprises deploy their software systems in the cloud environment. The cloud applications are usually large scale and include a lot of distributed cloud components. Building highly reliable cloud applications is a challenging and critical research problem. To attack this challenge, we propose a component ranking framework, named FTCloud, for building fault-tolerant cloud applications. FTCloud includes two ranking algorithms. The first algorithm employs component invocation structures and invocation frequencies for making significant component ranking. The second ranking algorithm systematically fuses the system structure information as well as the application designers' wisdom to identify the significant components in a cloud application. After the component ranking phase, an algorithm is proposed to automatically determine an optimal fault-tolerance strategy for the significant cloud components. The experimental results show that by tolerating faults of a small part of the most significant components, the reliability of cloud applications can be greatly improved.

Expertise retrieval, whose task is to suggest people with relevant expertise on the topic of interest, has received increasing interest in recent years. One of the issues is that previous algorithms mainly consider the documents associated with the experts while ignoring the community information that is affiliated with the documents and the experts. Motivated by the observation that communities could provide valuable insight and distinctive information, we investigate and develop two community-aware strategies to enhance expertise retrieval. We first propose a new smoothing method using the community context for statistical language modeling, which is employed to identify the most relevant documents so as to boost the performance of expertise retrieval in the document-based model. Furthermore, we propose a query-sensitive AuthorRank to model the authors' authorities based on the community coauthorship networks and develop an adaptive ranking refinement method to enhance expertise retrieval. Experimental results demonstrate the effectiveness and robustness of both community-aware strategies. Moreover, the improvements made in the enhanced models are significant and consistent.

Strongly promoted by the leading industrial companies, cloud computing becomes increasingly popular in re-cent years. The growth rate of cloud computing surpasses even the most optimistic predictions. A cloud application is a large-scale distributed system that consist a lot of distributed cloud nodes. How to make optimal deployment of cloud applications is a challenging research problem. When deploying a cloud application to the cloud environment, cloud node ranking is one of the most important approaches for selecting optimal cloud nodes for the cloud application. Traditional ranking methods usually rank the cloud nodes based on their QoS values, without considering the communication performance between cloud nodes. However, such kind of node relationship is very important for the communication-intensive cloud applications (e.g., Message Passing Interface (MPI) programs), which have a lot of communications between the selected cloud nodes. In this paper, we propose a novel clustering-based method for selecting optimal cloud nodes for deploying communication-intensive applications to the cloud environment. Our method not only takes into account the cloud node qualities, but also the communication performance between different nodes. We deploy several well-known MPI programs on a real-world cloud and compare our method with other methods. The experimental results show the effectiveness of our cluster-based method.

Cloud computing has attracted much interest recently from both industry and academic. Nowadays, more and more Internet applications are moving to the cloud environment. Making optimal deployment of cloud applications is critical for providing good performance to attract users. Optimizing user experience is usually required for cloud service deployment. However, it is a challenging task to know the user experience of end users, since there is generally no proactive connection between a user to the machine that will host the service instance. To attack this challenge, in this paper, we first propose a framework to model cloud features and capture user experience. Then based on the collected user connection information, we formulate the redeployment of service instances as k-median and max k-cover problems. We proposed several approximation algorithms to efficiently solve these problems. Comprehensive experiments are conducted by employing a real-world QoS dataset of service invocation. The experimental results show the effectiveness of our proposed redeployment approaches.

Cloud computing is becoming a popular and important solution for building highly reliable applications on distributed resources. However, it is a critical challenge to guarantee the system reliability of applications especially in voluntary-resource cloud due to the highly dynamic environment. In this paper, we present BFT Cloud (Byzantine Fault Tolerant Cloud), a Byzantine fault tolerance framework for building robust systems involuntary-resource cloud environments. BFT Cloud guarantees robustness of systems when up to f of totally 3f+1 resource providers are faulty, including crash faults, arbitrary behaviors faults, etc. BFT Cloud is evaluated in a large-scale real-world experiment which consists of 257 voluntary-resource providers located in 26 countries. The experimental results shows that BFT Cloud guarantees high reliability of systems built on the top of voluntary-resource cloud infrastructure and ensures good performance of these systems.

Online forums contain interactive and semantically related discussions on various questions. Extracted question-answer archive is invaluable knowledge, which can be used to improve Question Answering services. In this paper, we address the problem of Question Suggestion, which targets at suggesting questions that are semantically related to a queried question. Existing bag-of-words approaches suffer from the shortcoming that they could not bridge the lexical chasm between semantically related questions. Therefore, we present a new framework to suggest questions, and propose the Topicenhanced Translation-based Language Model (TopicTRLM) which fuses both the lexical and latent semantic knowledge. Extensive experiments have been conducted with a large real world data set. Experimental results indicate our approach is very effective and outperforms other popular methods in several metrics.

This paper investigates a ground-breaking incorporation of question category to Question Routing (QR) in Community Question Answering (CQA) services. The incorporation of question category was designed to estimate answerer expertise for routing questions to potential answerers. Two category-sensitive Language Models (LMs) were developed with large-scale real world data sets being experimented. Results demonstrated that higher accuracies of routing questions with lower computational costs were achieved, relative to traditional Query Likelihood LM (QLLM), state-of-the-art Cluster-Based LM (CBLM) and the mixture of Latent Dirichlet Allocation and QLLM (LDALM).

In this paper, we investigate the novel problem of automatic question identification in the microblog environment. It contains two steps: detecting tweets that contain questions (we call them "interrogative tweets") and extracting the tweets which really seek information or ask for help (so called "qweets") from interrogative tweets. To detect interrogative tweets, both traditional rule-based approach and state-of-the-art learning-based method are employed. To extract qweets, context features like short urls and Tweet-specific features like Retweets are elaborately selected for classification. We conduct an empirical study with sampled one hour's English tweets and report our experimental results for question identification on Twitter.

Previous semi-supervised learning (SSL) techniques usually assume unlabeled data are relevant to the target task. That is, they follow the same distribution as the targeted labeled data. In this paper, we address a different and very difficult scenario in SSL, where the unlabeled data may be a mixture of data relevant or irrelevant to the target binary classification task. In our framework, we do not require explicitly prior knowledge on the relatedness of the unlabeled data to the target data. In order to alleviate the effect of the irrelevant unlabeled data and utilize the implicit knowledge among all available data, we develop a novel maximum margin classifier, named the tri-class support vector machine (3C-SVM), to seek an inductive rule to separate the target binary classification task well while finding out the irrelevant data by-product. To attain this goal, we introduce a new min loss function, which can relieve the impact of the irrelevant data while relying more on the labeled data and the relevant unlabeled data. This loss function can therefore achieve the maximum entropy principle. The 3C-SVM can then generalize standard SVMs, Semi-supervised SVMs, and SVMs learned from the universum as its special cases. We further analyze the property of 3C-SVM on why the irrelevant data can help to improve the model performance. For implementation, we make relaxation and approximate the objective by the convex-concave procedure, which turns the original optimization from integral programming problem to a problem by just solving a finite number of quadratic programming problems. Empirical results are reported to demonstrate the advantages of our 3C-SVM model.

Most high-level information extraction (IE) consists of compound and aggregated subtasks. Such IE problems are generally challenging and they have generated increasing interest recently. We investigate two representative IE tasks: (1) entity identification and relation extraction from Wikipedia, and (2) citation matching, and we formally define joint optimization of information extraction. We propose a joint paradigm integrating three factors -- segmentation, relation, and segmentation-relation joint factors, to solve all relevant subtasks simultaneously. This modeling offers a natural formalism for exploiting bidirectional rich dependencies and interactions between relevant subtasks to capture mutual benefits. Since exact parameter estimation is prohibitively intractable, we present a general, highly-coupled learning algorithm based on variational expectation maximization (VEM) to perform parameter estimation approximately in a top-down and bottom-up manner, such that information can flow bidirectionally and mutual benefits from different subtasks can be well exploited. In this algorithm, both segmentation and relation are optimized iteratively and collaboratively using hypotheses from each other. We conducted extensive experiments using two real-world datasets to demonstrate the promise of our approach.

Since there are many Web services on the Internet, personalized Web service selection and recommendation is very important. In this paper, we present a new similarity measure for Web service similarity computation and propose a normal recovery collaborative filtering (NRCF) method for personalized Web service recommendation.

The exponential growth of Web service makes building high-quality service-oriented applications an urgent and crucial research problem. User-side QoS evaluations of Web services are critical for selecting the optimal Web service from a set of functionally equivalent service candidates. Since QoS performance of Web services is highly related to the service status and network environments which are variable against time, service invocations are required at different instances during a long time interval for making accurate Web service QoS evaluation. However, invoking a huge number of Web services from user-side for quality evaluation purpose is time-consuming, resource-consuming, and sometimes even impractical (e.g., service invocations are charged by service providers). To address this critical challenge, this paper proposes a Web service QoS prediction framework, called WSPred, to provide time-aware personalized QoS value prediction service for different service users. WSPred requires no additional invocation of Web services. Based on the past Web service usage experience from different service users, WSPred builds feature models and employs these models to make personalized QoS prediction for different users. The extensive experimental results show the effectiveness and efficiency of WSPred. Moreover, we publicly release our real-world time-aware Web service QoS dataset for future research, which makes our experiments verifiable and reproducible.

Web search engines have major impact in people's everyday life. It is of great importance to test the retrieval effectiveness of search engines. However, it is labor-intensive to judge the relevance of search results for a large number of queries, and these relevance judgments may not be reusable since the Web data change all the time. In this work, we propose to mine test oracles of Web search engines from existing search results. The main idea is to mine implicit relationships between queries and search results, e.g., some queries may have fixed top 1 result while some may not, and some Web domains may appear together in top 10 results. We define a set of items of queries and search results, and mine frequent association rules between these items as test oracles. Experiments on major search engines show that our approach mines many high-confidence rules that help understand search engines and detect suspicious search results.

Due to the prevalence of personalization and information filteringapplications, modelingusers ’ interests on theWeb has become increasingly important duringthe past few years. In this paper, aiming at providing accurate personalized Web site recommendations for Web users, we propose a novel probabilistic factor model based on dimensionality reduction techniques. We also extend the proposed method to collective probabilistic factor modeling, which further improves model performance by incorporating heterogeneous data sources. The proposed method is general, and can be applied to not only Web site recommendations, but also a wide range of Web applications, including behavioral targeting, sponsored search, etc. The experimental analysis on Web site recommendation shows that our method outperforms other traditional recommendation approaches. Moreover, the complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations. Categories and Subject Descriptors

Software systems are often built upon third party libraries. Developers may replace an old library with a new library, for the consideration of functionality, performance, security, and so on. It is tedious to learn the often complex APIs in the new library from the scratch. Instead, developers may identify the suitable APIs in the old library, and then find counterparts of these APIs in the new library. However, there is typically no such cross-references for APIs in different libraries. Previous work on automatic API recommendation often recommends related APIs in the same library. In this paper, we propose to mine search results of Web search engines to recommend related APIs of different libraries. In particular, we use Web search engines to collect relevant Web search results of a given API in the old library, and then recommend API candidates in the new library that are frequently appeared in the Web search results. Preliminary results of generating related C# APIs for the APIs in JDK show the feasibility of our approach.

With the increasing popularity of cloud computing as a solution for building high-quality applications on distributed components, efficiently evaluating user-side quality of cloud components becomes an urgent and crucial research problem. However, invoking all the available cloud components from user-side for evaluation purpose is expensive and impractical. To address this critical challenge, we propose a neighborhood-based approach, called CloudPred, for collaborative and personalized quality prediction of cloud components. CloudPred is enhanced by feature modeling on both users and components. Our approach CloudPred requires no additional invocation of cloud components on behalf of the cloud application designers. The extensive experimental results show that CloudPred achieves higher QoS prediction accuracy than other competing methods. We also publicly release our large-scale QoS dataset for future related research in cloud computing.

Despite the various applications of wireless sensor network (WSN), experiences from real WSN deployments show that protocol implementations in sensor nodes are susceptible to software failures, which may cause network failures or even breakdown. Pre-deployment protocol conformance testing is essential to ensure reliable communications for WSNs. Unfortunately, existing solutions with simulators cannot test the exact hardware and implementation environment as real sensors, whereas testbeds are expensive and limited to small scale networks and topologies. In this paper, we present RealProct, a novel and reliable framework for testing protocol implementations against their specifications in WSNs. RealProct utilizes real sensors for protocol conformance testing to ensure that the results are close to the real deployment. Using different techniques from those in simulations and real deployments, RealProct virtualizes a large network with any topology and generate non-deterministic events using only a small number of sensors to provide flexibility and to reduce the cost. The framework is carefully designed to support efficient testing in resource-limited sensors. Moreover, test execution and verdict are optimized to minimize the number of runs, while guaranteeing satisfactory false posi tive and false negative rates. We implement RealProct and test it with the IIP TCP/IP protocol stack and a routing protocol developed for WSNs in Contiki-2.4. The results demonstrate the effectiveness of RealProct by detecting several new bugs and all previously discovered bugs in various versions of the μIP TCP/IP protocol stack.

Although Recommender Systems have been comprehensively analyzed in the past decade, the study of social-based recommender systems just started. In this paper, aiming at providing a general method for improving recommender systems by incorporating social network information, we propose a matrix factorization framework with social regularization. The contributions of this paper are four-fold: (1) We elaborate how social network information can benefit recommender systems; (2) We interpret the differences between social-based recommender systems and trust-aware recommender systems; (3) We coin the term Social Regularization to represent the social constraints on recommender systems, and we systematically illustrate how to design a matrix factorization objective function with social regularization; and (4) The proposed method is quite general, which can be easily extended to incorporate other contextual information, like social tags, etc. The empirical analysis on two large datasets demonstrates that our approaches outperform other state-of-the-art methods.

The research issue of recommender systems has been treated as a classical regression problem over the decades and has obtained a great success. In the next generation of recommender systems, multi-criteria recommendation has been predicted as an important direction. Different from traditional recommender systems that aim particularly at recommending high-quality items evaluated by users' ratings, inmulti-criteria recommendation, quality only serves as one criterion, and many other criteria such as relevance, coverage, and diversity should be simultaneously optimized. Although recently there is work investigating each single criterion, there is rarely any literature that reports how each single criterion impacts each other and how to combine them in real applications. Thus in this paper, we study the relationship of two criteria, quality and relevance, as a preliminary work in multi-criteria recommendation. We first give qualitative and quantitative analysis of competitive quality-based and relevance-based algorithms in these two criteria to show that both algorithms cannot work well in the opposite criteria. Then we propose an integrated metric and finally investigate how to combine previous work together into an unified model. In the combination, we introduce a Continuous-time MArkov Process (CMAP) algorithm for ranking, which enables principled and natural integration with features derived from both quality-based and relevance-based algorithms. Through experimental verification, the combined methods can significantly outperform either single quality-based or relevance-based algorithms in the integrated metric and the CMAP model outperforms traditional combination methods by around 3%. Its linear complexity with respect to the number of users and items leads to satisfactory performance, as demonstrated by the around 7-hour computational time for over 480k users and almost 20k items.

Recommender systems have been well studied and developed, both in academia and in industry recently. However, traditional recommender systems assume that all the users are independent and identically distributed; this assumption ignores the connections among users, which is not consistent with the real-world observations where we always turn to our trusted friends for recommendations. Aiming at modeling recommender systems more accurately and realistically, we propose a novel probabilistic factor analysis framework which naturally fuses the users' tastes and their trusted friends' favors together. The proposed framework is quite general, and it can also be applied to pure user-item rating matrix even if we do not have explicit social trust information among users. In this framework, we coin the term social trust ensemble to represent the formulation of the social trust restrictions on the recommender systems. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations, while the experimental results show that our method outperforms state-of-the-art approaches.

Kernel methods have been successfully applied in various applications. To succeed in these applications, it is crucial to learn a good kernel representation, whose objective is to reveal the data similarity precisely. In this paper, we address the problem of multiple kernel learning (MKL), searching for the optimal kernel combination weights through maximizing a generalized performance measure. Most MKL methods employ the L(1)-norm simplex constraints on the kernel combination weights, which therefore involve a sparse but non-smooth solution for the kernel weights. Despite the success of their efficiency, they tend to discard informative complementary or orthogonal base kernels and yield degenerated generalization performance. Alternatively, imposing the L(p)-norm (p > 1) constraint on the kernel weights will keep all the information in the base kernels. This leads to non-sparse solutions and brings the risk of being sensitive to noise and incorporating redundant information. To tackle these problems, we propose a generalized MKL (GMKL) model by introducing an elastic-net-type constraint on the kernel weights. More specifically, it is an MKL model with a constraint on a linear combination of the L(1)-norm and the squared L(2)-norm on the kernel weights to seek the optimal kernel combination weights. Therefore, previous MKL problems based on the L(1)-norm or the L(2)-norm constraints can be regarded as special cases. Furthermore, our GMKL enjoys the favorable sparsity property on the solution and also facilitates the grouping effect. Moreover, the optimization of our GMKL is a convex optimization problem, where a local solution is the global optimal solution. We further derive a level method to efficiently solve the optimization problem. A series of experiments on both synthetic and real-world datasets have been conducted to show the effectiveness and efficiency of our GMKL.

Due to their potential commercial value and the associated great research challenges, recommender systems have been extensively studied by both academia and industry recently. However, the data sparsity problem of the involved user-item matrix seriously affects the recommendation quality. Many existing approaches to recommender systems cannot easily deal with users who have made very few ratings. In view of the exponential growth of information generated by online users, social contextual information analysis is becoming important for many Web applications. In this article, we propose a factor analysis approach based on probabilistic matrix factorization to alleviate the data sparsity and poor prediction accuracy problems by incorporating social contextual information, such as social networks and social tags. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations. Moreover, the experimental results show that our method performs much better than the state-of-the-art approaches, especially in the circumstance that users have made few ratings.

Near-duplicate keyframe (NDK) retrieval techniques are critical to many real-world multimedia applications. Over the last few years, we have witnessed a surge of attention on studying near-duplicate image/keyframe retrieval in the multimedia community. To facilitate an effective approach to NDK retrieval on large-scale data, we suggest an effective Multi-Level Ranking (MLR) scheme that effectively retrieves NDKs in a coarse-to-fine manner. One key stage of the MLR ranking scheme is how to learn an effective ranking function with extremely small training examples in a near-duplicate detection task. To attack this challenge, we employ a semi-supervised learning method, semi-supervised support vector machines, which is able to significantly improve the retrieval performance by exploiting unlabeled data. Another key stage of the MLR scheme is to perform a fine matching among a subset of keyframe candidates retrieved from the previous coarse ranking stage. In contrast to previous approaches based on either simple heuristics or rigid matching models, we propose a novel Nonrigid Image Matching (NIM) approach to tackle near-duplicate keyframe retrieval from real-world video corpora in order to conduct an effective fine matching. Compared with the conventional methods, the proposed NIM approach can recover explicit mapping between two near-duplicate images with a few deformation parameters and find out the correct correspondences from noisy data simultaneously. To evaluate the effectiveness of our proposed approach, we performed extensive experiments on two benchmark testbeds extracted from the TRECVID2003 and TRECVID2004 corpora. The promising results indicate that our proposed method is more effective than other state-of-the-art approaches for near-duplicate keyframe retrieval.

Software typically undergoes debugging during both a testing phase before product release, and an operational phase after product release. But it is noted that the fault detection and removal processes during software development and operation are different. For example, the fault removal during operation occurs generally at a slower pace than development. In this paper, we derive a powerful, easily deployable technique for software reliability prediction and assessment in the testing and operational phases. We first review how several existing software reliability growth models (SRGM) based on non- homogeneous Poisson processes (NHPP) can be readily derived from a unified theory. With the unified theory, we further incorporate the concept of multiple change-points, i.e. points in time when the software environment changes, into software reliability modeling. Several models are proposed and discussed under both ideal and imperfect debugging conditions. We estimate the parameters of the proposed models by employing real software failure data, and give a fair comparison with some existing SRGM. Numerical results show that the proposed models can provide good software reliability prediction in the various stages of software development and operation. Our approach is flexible; we can model various environments ranging from exponential-type to S-shaped NHPP models.

With increasing presence and adoption of Web services on the World Wide Web, Quality-of-Service (QoS) is becoming important for describing nonfunctional characteristics of Web services. In this paper, we present a collaborative filtering approach for predicting QoS values of Web services and making Web service recommendation by taking advantages of past usage experiences of service users. We first propose a user-collaborative mechanism for past Web service QoS information collection from different service users. Then, based on the collected QoS data, a collaborative filtering approach is designed to predict Web service QoS values. Finally, a prototype called WSRec is implemented by Java language and deployed to the Internet for conducting real-world experiments. To study the QoS value prediction accuracy of our approach, 1.5 millions Web service invocation results are collected from 150 service users in 24 countries on 100 real-world Web services in 22 countries. The experimental results show that our algorithm achieves better prediction accuracy than other approaches. Our Web service QoS data set is publicly released for future research.

In order to improve the user search experience, Query Suggestion, a technique for generating alternative queries to Web users, has become an indispensable feature for commercial search engines. However, previous work mainly focuses on suggesting relevant queries to the original query while ignoring the diversity in the suggestions, which will potentially dissatisfy Web users' information needs. In this paper, we present a novel unified method to suggest both semantically relevant and diverse queries to Web users. The proposed approach is based on Markov random walk and hitting time analysis on the query-URL bipartite graph. It can effectively prevent semantically redundant queries from receiving a high rank, hence encouraging diversities in the results. We evaluate our method on a large commercial clickthrough dataset in terms of relevance measurement and diversity measurement. The experimental results show that our method is very effective in generating both relevant and diverse query suggestions.

Multiple Kernel Learning (MKL) can be formulated as a convex-concave minmax optimization problem, whose saddle point corresponds to the optimal solution to MKL. Most MKL methods employ the L1-norm simplex constraints on the combination weights of kernels, which therefore involves optimization of a non-smooth function of the kernel weights. These methods usually divide the optimization into two cycles: one cycle deals with the optimization on the kernel combination weights, and the other cycle updates the parameters of SVM. Despite the success of their efficiency, they tend to discard informative complementary kernels. To improve accuracy, we introduce smoothness to the optimization procedure. Furthermore, we transform the optimization into a single smooth convex optimization problem and employ the Nesterov’s method to efficiently solve the optimization problem. Experiments on benchmark data sets demonstrate that the proposed algorithm clearly improves current MKL methods in a number scenarios.

Social tagging systems have emerged as an effective way for users to annotate and share objects on the Web. However, with the growth of social tagging systems, users are easily overwhelmed by the large amount of data and it is very difficult for users to dig out information that he/she is interested in. Though the tagging system has provided interest-based social network features to enable the user to keep track of other users' tagging activities, there is still no automatic and effective way for the user to discover other users with common interests. In this paper, we propose a User Recommendation (UserRec) framework for user interest modeling and interest-based user recommendation, aiming to boost information sharing among users with similar interests. Our work brings three major contributions to the research community: (1) we propose a tag-graph based community detection method to model the users' personal interests, which are further represented by discrete topic distributions; (2) the similarity values between users' topic distributions are measured by Kullback-Leibler divergence (KL-divergence), and the similarity values are further used to perform interest-based user recommendation; and (3) by analyzing users' roles in a tagging system, we find users' roles in a tagging system are similar to Web pages in the Internet. Experiments on tagging dataset of Web pages (Yahoo!~Delicious) show that UserRec outperforms other state-of-the-art recommender system approaches.

Multi-task feature selection (MTFS) is an important tool to learn the explanatory features across multiple related tasks. Previous MTFS methods fulfill this task in batch-mode training. This makes them inefficient when data come in sequence or when the number of training data is so large that they cannot be loaded into the memory simultaneously. To tackle these problems, we propose the first online learning framework for MTFS. A main advantage of the online algorithms is the efficiency in both time complexity and memory cost due to the closed-form solutions in updating the model weights at each iteration. Experimental results on a real-world dataset attest to the merits of the proposed algorithms.

Wireless Sensor Network (WSN) applications are typically event-driven. While the source codes of these applications may look simple, they are executed with a complicated concurrency model, which frequently introduces software bugs, in particular, transient bugs. Such buggy logics may only be triggered by some occasionally interleaved events that bear implicit dependency, but can lead to fatal system failures. Unfortunately, these deeply-hidden bugs or even their symptoms can hardly be identified by state-of-the-art debugging tools, and manual identification from massive running traces can be prohibitively expensive. In this paper, we present Sentomist (Sensor application anatomist), a novel tool for identifying potential transient bugs in WSN applications. The Sentomist design is based on a key observation that transient bugs make the behaviors of a WSN system deviate from the normal, and thus outliers (i.e., abnormal behaviors) are good indicators of potential bugs. Sentomist introduces the notion of event-handling interval to systematically anatomize the long-term execution history of an event-driven WSN system into groups of intervals. It then applies a customized outlier detection algorithm to quickly identify and rank abnormal intervals. This dramatically reduces the human efforts of inspection (otherwise, we have to manually check tremendous data samples, typically with brute force inspection) and thus greatly speeds up debugging. We have implemented Sentomist based on the concurrency model of TinyOS. We apply Sentomist to test a series of representative real-life WSN applications that contain transient bugs. These bugs, though caused by complicated interactions that can hardly be predicted during the programming stage, are successfully confined by Sentomist.

Service-oriented architecture (SOA) is becoming a major software framework for building complex distributed systems. Reliability of the service-oriented systems heavily depends on the remote Web services as well as the unpredictable Internet. Designing effective and accurate reliability prediction approaches for the service-oriented systems has become an important research issue. In this paper, we propose a collaborative reliability prediction approach, which employs the past failure data of other similar users to predict the Web service reliability for the current user, without requiring real-world Web service invocations. We also present a user-collaborative failure data sharing mechanism and a reliability composition model for the service-oriented systems. Large-scale real-world experiments are conducted and the experimental results show that our collaborative reliability prediction approach obtains better reliability prediction accuracy than other approaches.

Web services are becoming prevalent nowadays. Finding desired Web services is becoming an emergent and challenging research problem. In this paper, we present WSExpress (Web Service Express), a novel Web service search engine to expressively find expected Web services. WSExpress ranks the publicly available Web services not only by functional similarities to users' queries, but also by nonfunctional QoS characteristics of Web services. WSExpress provides three searching styles, which can adapt to the scenario of finding an appropriate Web service and the scenario of automatically replacing a failed Web service with a suitable one. WSExpress is implemented by Java language and large-scale experiments employing real-world Web services are conducted. Totally 3,738 Web services (15,811 operations) from 69 countries are involved in our experiments. The experimental results show that our search engine can find Web services with the desired functional and non-functional requirements. Extensive experimental studies are also conducted on a well known benchmark dataset consisting of 1,000 Web service operations to show the recall and precision performance of our search engine.

Quality-of-Service (QoS) is widely employed for describing non-functional characteristics of Web services. Although QoS of Web services has been investigated in a lot of previous works, there is a lack of real-world Web service QoS datasets for validating new QoS based techniques and models of Web services. To study the performance of real-world Web services as well as provide reusable research datasets for promoting the research of QoS-driven Web services, we conduct several large-scale evaluations on real-world Web services. Firstly, addresses of 21,358 Web services are obtained from the Internet. Then, invocation failure probability performance of 150 Web services is assessed by 100 distributed service users. After that, response time and throughput performance of 5,825 Web services are evaluated by 339 distributed service users. Detailed experimental results are presented in this paper and comprehensive Web service QoS datasets are publicly released for future research.

In this paper, we address the problem of one-class classification. Taking into account the fact that in some applications, the given training samples are rather limited, we attempt to utilize the advantages of Multi-task Learning (MTL), where the data of related tasks may share similar structure and helpful information. We then propose an MTL framework for one-class classification. The framework derives from the one-class v-SVM and makes use of related tasks by constraining them to have similar solutions. This formulation can be cast into a second-order cone program, which achieves a global solution and is solved efficiently. Further, the framework also maintains the favorable property of the v parameter in the v-SVM, which can control the fraction of outliers and support vectors, in one-class classification. This framework also connects with several existing models. Experimental results on both synthetic and real-world datasets demonstrate the properties and advantages of our proposed model.

Cloud computing is becoming a mainstream aspect of information technology. The cloud applications are usually large-scale, complex, and include a lot of distributed components. Providing highly reliable cloud applications is a challenging and critical research problem. To attack this challenge, we propose FTCloud which is a component ranking based framework for building fault-tolerant cloud applications. FTCloud employs the component invocation structures and the invocation frequencies to identify the significant components in a cloud application. An algorithm is proposed to automatically determine optimal fault tolerance strategy for these significant components. The experimental results show that by tolerating faults of a small part of the most significant components, the reliability of cloud application can be greatly improved.

A key component of automated object-oriented unit-test generation is to find method-call sequences that generate desired inputs of a method under test (MUT). Previous work cannot find desired sequences effectively due to the large search space of possible sequences. To address this issue, we present a MUT-aware sequence recommendation approach called RecGen to improve the effectiveness of random object-oriented unit-test generation. Unlike existing random testing approaches that select sequences without considering how a MUT may use inputs generated from sequences, RecGen analyzes object fields accessed by a MUT and recommends a short sequence that mutates these fields. In addition, for MUTs whose test generation keeps failing, RecGen recommends a set of sequences to cover all the methods that mutate object fields accessed by the MUT. This technique further improves the chance of generating desired inputs. We have implemented RecGen and evaluated it on three libraries. Evaluation results show that RecGen improves code coverage over previous random testing tools.

Underwater acoustic sensor networks (UWASNs) are composed of underwater sensors that use sound to transmit information collected in the ocean. Since the sound speed is lower than radio wave, UWASNs suffer from much lower throughput and higher delay compared with terrestrial wireless sensor networks (TWSNs). Current methods manage to alleviate the bottleneck by replacing mutual handshakes with reservation mechanisms that consume lower overhead. However, their throughput improvement and delay reduction are very limited (e.g., the throughput is only 30% of the theoretical maximum for TLohi in 8-node networks), and most of their analysis and simulations are based on single-hop communication. In this work, we tackle the above challenges by proposing a priority scheduling approach for multi-hop topologies. First, we find that the scheduling problem of UWASNs is very different from that of TWSNs, and analyze the shortest schedule for the whole UWASN. Then, we design an efficient priority scheduling protocol at the MAC layer. Our approach performs parallel transmissions and prioritizes the scheduling by allocating longer time to heavier-traffic nodes. We have conducted extensive evaluations which show that the proposed protocol not only improves the throughput and delay performance greatly, but also benefits the fairness.

The rising popularity of cloud computing makes building high quality cloud applications a critical and urgently required research problem. Component quality ranking approaches are crucial for making optimal component selection from a set of functionally equivalent component candidates. Moreover, quality ranking of cloud components helps the application designers detect the poor performing components in the complex cloud applications, which usually include huge number of distributed components. To provide personalized cloud component ranking for different designers of cloud applications, this paper proposes a QoS-driven component ranking framework for cloud applications by taking advantage of the past component usage experiences of different component users. Our approach requires no additional invocations of the cloud components on behalf of the application designers. The extensive experimental results show that our approach outperforms the competing approaches.

As the exponential growth of information generated on the World Wide Web, Social Recommendation has emerged as one of the hot research topics recently. Social Recommendation forms a specific type of information filtering technique that attempts to suggest information (blogs, news, music, travel plans, web pages, images, tags, etc.) that are likely to interest the users. Social Recommendation involves the investigation of collective intelligence by using computational techniques such as machine learning, data mining, natural language processing, etc. on social behavior data collected from blogs, wikis, recommender systems, question & answer communities, query logs, tags, etc. from areas such as social networks, social search, social media, social bookmarks, social news, social knowledge sharing, and social games. In this tutorial, we will introduce Social Recommendation and elaborate on how the various characteristics and aspects are involved in the social platforms for collective intelligence. Moreover, we will discuss the challenging issues involved in Social Recommendation in the context of theory and models of social networks, methods to improve recommender systems using social contextual information, ways to deal with partial and incomplete information in the social context, scalability and algorithmic issues with social computational techniques.

Abstract Domain testing is designed to detect domain errors that result from a small boundary shift in a path domain. Although many researchers have studied domain testing, automatic domain test data generation for string predicates has seldom been explored. This paper presents a novel approach for the automatic generation ofON–OFFtest points for string predicate borders, and describes a corresponding test data generator. Our empirical work is conducted on a set of programs with string predicates, where extensive trials have been done for each string predicate, and the results are analysed using the SPSS tool. Conclusions are drawn that: (i) the approach is promising and effective; (ii) there is a strong linear relationship between the performance of the test generator and the length of target string in the predicate tested; and (iii) initial inputs, no shorter than the target string and with characters generated randomly, may enhance the performance in the test data generation for string predicates. Copyright © 2009 John Wiley & Sons, Ltd.

With the exponential growth of Web 2.0 applications, tags have been used extensively to describe the image contents on the Web. Due to the noisy and sparse nature in the human generated tags, how to understand and utilize these tags for image retrieval tasks has become an emerging research direction. As the low-level visual features can provide fruitful information, they are employed to improve the image retrieval results. However, it is challenging to bridge the semantic gap between image contents and tags. To attack this critical problem, we propose a unified framework in this paper which stems from a two-level data fusions between the image contents and tags: 1) A unified graph is built to fuse the visual feature-based image similarity graph with the image-tag bipartite graph; 2) A novel random walk model is then proposed, which utilizes a fusion parameter to balance the influences between the image contents and tags. Furthermore, the presented framework not only can naturally incorporate the pseudo relevance feedback process, but also it can be directly applied to applications such as content-based image retrieval, text-based image retrieval, and image annotation. Experimental analysis on a large Flickr dataset shows the effectiveness and efficiency of our proposed framework.

Feature selection has attracted a huge amount of interest in both research and application communities of data mining. We consider the problem of semi-supervised feature selection, where we are given a small amount of labeled examples and a large amount of unlabeled examples. Since a small number of labeled samples are usually insufficient for identifying the relevant features, the critical problem arising from semi-supervised feature selection is how to take advantage of the information underneath the unlabeled data. To address this problem, we propose a novel discriminative semi-supervised feature selection method based on the idea of manifold regularization. The proposed approach selects features through maximizing the classification margin between different classes and simultaneously exploiting the geometry of the probability distribution that generates both labeled and unlabeled data. In comparison with previous semi-supervised feature selection algorithms, our proposed semi-supervised feature selection method is an embedded feature selection method and is able to find more discriminative features. We formulate the proposed feature selection method into a convex-concave optimization problem, where the saddle point corresponds to the optimal solution. To find the optimal solution, the level method, a fairly recent optimization method, is employed. We also present a theoretic proof of the convergence rate for the application of the level method to our problem. Empirical evaluation on several benchmark data sets demonstrates the effectiveness of the proposed semi-supervised feature selection method.

Expertise retrieval has received increased interests in recent years, whose task is to suggest people with relevant expertise. Motivated by the observation that communities could provide valuable insight and distinctive information, we investigate two community-aware strategies to enhance expertise retrieval. We first propose a new smoothing method using the community context instead of the whole collection for statistical language model in the document-based model. Furthermore, a query-sensitive AuthorRank is proposed to model the authors' authorities according to the community co-authorship networks, and then an adaptive ranking refinement method is developed to further enhance expertise retrieval. Experimental results demonstrate the effectiveness and robustness of both community-aware strategies.

The problem of measuring similarity between web pages arises in many important Web applications, such as search engines and Web directories. In this paper, we propose a novel neighbor-based similarity measure called MatchSim, which uses only the neighborhood structure of web pages. Technically, MatchSim recursively defines similarity between web pages by the average similarity of the maximum matching between their neighbors. Our method extends the traditional methods which simply count the numbers of common and/or different neighbors. It also successfully overcomes a severe counterintuitive loophole in SimRank, due to its strict consistency with the intuitions of similarity. We give the computational complexity of MatchSim iteration. The accuracy of MatchSim is compared against others on two real datasets. The results show that our method performs best in most cases.

Collaborative Filtering, considered by many researchers as the most important technique for information filtering, has been extensively studied by both academic and industrial communities. One of the most popular approaches to collaborative filtering recommendation algorithms is based on low-dimensional factor models. The assumption behind such models is that a user's preferences can be modeled by linearly combining item factor vectors using user-specific coefficients. In this paper, aiming at several aspects ignored by previous work, we propose a semi-nonnegative matrix factorization method with global statistical consistency. The major contribution of our work is twofold: (1) We endow a new understanding on the generation or latent compositions of the user-item rating matrix. Under the new interpretation, our work can be formulated as the semi-nonnegative matrix factorization problem. (2) Moreover, we propose a novel method of imposing the consistency between the statistics given by the predicted values and the statistics given by the data. We further develop an optimization algorithm to determine the model complexity automatically. The complexity of our method is linear with the number of the observed ratings, hence it is scalable to very large datasets. Finally, comparing with other state-of-the-art methods, the experimental analysis on the EachMovie dataset illustrates the effectiveness of our approach.

This paper addresses the issue of social recommendation based on collaborative filtering (CF) algorithms. Social recommendation emphasizes utilizing various attributes information and relations in social networks to assist recommender systems. Although recommendation techniques have obtained distinct developments over the decades, traditional CF algorithms still have these following two limitations: (1) relational dependency within predictions, an important factor especially when the data is sparse, is not being utilized effectively; and (2) straightforward methods for combining features like linear integration suffer from high computing complexity in learning the weights by enumerating the whole value space, making it difficult to combine various information into an unified approach. In this paper, we propose a novel model, Multi-scale Continuous Conditional Random Fields (MCCRF), as a framework to solve above problems for social recommendations. In MCCRF, relational dependency within predictions is modeled by the Markov property, thus predictions are generated simultaneously and can help each other. This strategy has never been employed previously. Besides, diverse information and relations in social network can be modeled by state and edge feature functions in MCCRF, whose weights can be optimized globally. Thus both problems can be solved under this framework. In addition, We propose to utilize Markov chain Monte Carlo (MCMC) estimation methods to solve the difficulties in training and inference processes of MCCRF. Experimental results conducted on two real world data have demonstrated that our approach outperforms traditional CF algorithms. Additional experiments also show the improvements from the two factors of relational dependency and feature combination, respectively.

Most state-of-the-art nonrigid shape recovery methods usually use explicit deformable mesh models to regularize surface deformation and constrain the search space. These triangulated mesh models heavily relying on the quadratic regularization term are difficult to accurately capture large deformations, such as severe bending. In this paper, we propose a novel Gaussian process regression approach to the nonrigid shape recovery problem, which does not require to involve a predefined triangulated mesh model. By taking advantage of our novel Gaussian process regression formulation together with a robust coarse-to-fine optimization scheme, the proposed method is fully automatic and is able to handle large deformations and outliers. We conducted a set of extensive experiments for performance evaluation in various environments. Encouraging experimental results show that our proposed approach is both effective and robust to nonrigid shape recovery with large deformations.

Based on the framework of service-oriented architecture (SOA), complex distributed systems can be dynamically and automatically composed by integrating distributed Web services provided by different organizations, making dependability of the distributed SOA systems a big challenge. In this paper, we propose a QoS-aware fault tolerant middleware to attack this critical problem. Our middleware includes a user-collaborated QoS model, various fault tolerance strategies, and a context-aware algorithm in determining optimal fault tolerance strategy for both stateless and stateful Web services. The benefits of the proposed middleware are demonstrated by experiments, and the performance of the optimal fault tolerance strategy selection algorithm is investigated extensively. As illustrated by the experimental results, fault tolerance for the distributed SOA systems can be efficient, effective and optimized by the proposed middleware.

We consider the problem of selecting a subset of m most informative features where m is the number of required features. This feature selection problem is essentially a combinatorial optimization problem, and is usually solved by an approximation. Conventional feature selection methods address the computational challenge in two steps: (a) ranking all the features by certain scores that are usually computed independently from the number of specified features m, and (b) selecting the top m ranked features. One major shortcoming of these approaches is that if a feature f is chosen when the number of specified features is m, it will always be chosen when the number of specified features is larger than m. We refer to this property as the "monotonic" property of feature selection. In this work, we argue that it is important to develop efficient algorithms for non-monotonic feature selection. To this end, we develop an algorithm for non-monotonic feature selection that approximates the related combinatorial optimization problem by a Multiple Kernel Learning (MKL) problem. We also present a strategy that derives a discrete solution from the approximate solution of MKL, and show the performance guarantee for the derived discrete solution when compared to the global optimal solution for the related combinatorial optimization problem. An empirical study with a number of benchmark data sets indicates the promising performance of the proposed framework compared with several state-of-the-art approaches for feature selection.

As the abundance of Web services on the World Wide Web increase, designing effective approaches for Web service selection and recommendation has become more and more important. In this paper, we present WSRec, a Web service recommender system, to attack this crucial problem. WSRec includes a user-contribution mechanism for Web service QoS information collection and an effective and novel hybrid collaborative filtering algorithm for Web service QoS value prediction. WSRec is implemented by Java language and deployed to the real-world environment. To study the prediction performance, a total of 21,197 public Web services are obtained from the Internet and a large-scale real-world experiment is conducted, where more than 1.5 millions test results are collected from 150 service users in different countries on 100 publicly available Web services located all over the world. The comprehensive experimental analysis shows that WSRec achieves better prediction accuracy than other approaches.

We consider the task of Self-taught Learning (STL) from unlabeled data. In contrast to semi-supervised learning, which requires unlabeled data to have the same set of class labels as labeled data, STL can transfer knowledge from different types of unlabeled data. STL uses a three-step strategy: (1) learning high-level representations from unlabeled data only, (2) re-constructing the labeled data via such representations and (3) building a classifier over the re-constructed labeled data. However, the high-level representations which are exclusively determined by the unlabeled data, may be inappropriate or even misleading for the latter classifier training process. In this paper, we propose a novel Supervised Self-taught Learning (SSTL) framework that successfully integrates the three isolated steps of STL into a single optimization problem. Benefiting from the interaction between the classifier optimization and the process of choosing high-level representations, the proposed model is able to select those discriminative representations which are more appropriate for classification. One important feature of our novel framework is that the final optimization can be iteratively solved with convergence guaranteed. We evaluate our novel framework on various data sets. The experimental results show that the proposed SSTL can outperform STL and traditional supervised learning methods in certain instances.

Recently many data types arising from data mining and Web search applications can be modeled as bipartite graphs. Examples include queries and URLs in query logs, and authors and papers in scientific literature. However, one of the issues is that previous algorithms only consider the content and link information from one side of the bipartite graph. There is a lack of constraints to make sure the final relevance of the score propagation on the graph, as there are many noisy edges within the bipartite graph. In this paper, we propose a novel and general Co-HITS algorithm to incorporate the bipartite graph with the content information from both sides as well as the constraints of relevance. Moreover, we investigate the algorithm based on two frameworks, including the iterative and the regularization frameworks, and illustrate the generalized Co-HITS algorithm from different views. For the iterative framework, it contains HITS and personalized PageRank as special cases. In the regularization framework, we successfully build a connection with HITS, and develop a new cost function to consider the direct relationship between two entity sets, which leads to a significant improvement over the baseline method. To illustrate our methodology, we apply the Co-HITS algorithm, with many different settings, to the application of query suggestion by mining the AOL query log data. Experimental results demonstrate that CoRegu-0.5 (i.e., a model of the regularization framework) achieves the best performance with consistent and promising improvements.

Geographic forwarding is a favorable scheme for data reporting in wireless sensor networks (WSNs) due to its simplicity and low-overhead. However, WSNs are usually subject to complicated environmental factors. Network holes (i.e., the areas where no nodes inside) and barriers (i.e., those blocking the communication between two close nodes) are inevitable in practical deploying environments. These issues pose an obstacle to adopting geographic forwarding in WSNs, while current approaches lack an efficient method to tolerate such negative factors. In this paper we specifically tailor a waypoint-based geographic data reporting protocol (GDRP) for WSNs. Inherited from geographic forwarding, GDRP is light-weighted and hence well-suits WSNs. But unlike current approaches that often find suboptimal paths, GDRP adopts an intelligent strategy to select a best set of waypoints via which packets can efficiently circumvent holes and barriers, and it can thus find better paths. Extensive simulations are conducted to verify the advantages of GDRP in tolerating network holes and obstacles in WSNs.

Contour mapping is an important technique for Wireless Sensor Networks (WSNs) in environmental monitoring to abstract the information of a monitored field. State-of-the-art approaches for contour mapping, however, are neither energy-optimized, nor capable of handling heterogeneous user requests. In this paper, we develop a novel energy-efficient On-demand Active Contour Service (OACS) for power-constrained WSNs. OACS regresses the field intensity function with kernel Support Vector Regression (SVR), a novel machine learning tool that flexibly handles both contour line and contour map requests. OACS also adaptively accommodates a wide range of contour line/map precision requirements: (1) For applications of low precision, only a minimum set of nodes are scheduled in working mode while others are sleeping for conserving energy. (2) For applications of high precision, through an active and progressive learning algorithm, OACS determines the best set of nodes that should be turned on for improving the contour line/map precision. Evaluation based on diverse realistic models demonstrates that OACS provides quality and seamless contour services for various application requirements yet significantly conserves energy.

We discuss the framework of Transductive Support Vector Machine (TSVM) from the perspective of the regularization strength induced by the unlabeled data. In this framework, SVM and TSVM can be regarded as a learning machine without regularization and one with full regularization from the unlabeled data, respectively. Therefore, to supplement this framework of the regularization strength, it is necessary to introduce data-dependant partial regularization. To this end, we reformulate TSVM into a form with controllable regularization strength, which includes SVM and TSVM as special cases. Furthermore, we introduce a method of adaptive regularization that is data dependant and is based on the smoothness assumption. Experiments on a set of benchmark data sets indicate the promising results of the proposed work compared with state-of-the-art TSVM algorithms.

With the exponential growth of Web contents, Recommender System has become indispensable for discovering new information that might interest Web users. Despite their success in the industry, traditional recommender systems suffer from several problems. First, the sparseness of the user-item matrix seriously affects the recommendation quality. Second, traditional recommender systems ignore the connections among users, which loses the opportunity to provide more accurate and personalized recommendations. In this paper, aiming at providing more realistic and accurate recommendations, we propose a factor analysis-based optimization framework to incorporate the user trust and distrust relationships into the recommender systems. The contributions of this paper are three-fold: (1) We elaborate how user distrust information can benefit the recommender systems. (2) In terms of the trust relations, distinct from previous trust-aware recommender systems which are based on some heuristics, we systematically interpret how to constrain the objective function with trust regularization. (3) The experimental results show that the distrust relations among users are as important as the trust relations. The complexity analysis shows our method scales linearly with the number of observations, while the empirical analysis on a large Epinions dataset proves that our approaches perform better than the state-of-the-art approaches.

Query log analysis has received substantial attention in recent years, in which the click graph is an important technique for describing the relationship between queries and URLs. State-of-the-art approaches based on the raw click frequencies for modeling the click graph, however, are not noise-eliminated. Nor do they handle heterogeneous query-URL pairs well. In this paper, we investigate and develop a novel entropy-biased framework for modeling click graphs. The intuition behind this model is that various query-URL pairs should be treated differently, i.e., common clicks on less frequent but more specific URLs are of greater value than common clicks on frequent and general URLs. Based on this intuition, we utilize the entropy information of the URLs and introduce a new concept, namely the inverse query frequency (IQF), to weigh the importance (discriminative ability) of a click on a certain URL. The IQF weighting scheme is never explicitly explored or statistically examined for any bipartite graphs in the information retrieval literature. We not only formally define and quantify this scheme, but also incorporate it with the click frequency and user frequency information on the click graph for an effective query representation. To illustrate our methodology, we conduct experiments with the AOL query log data for query similarity analysis and query suggestion tasks. Experimental results demonstrate that considerable improvements in performance are obtained with our entropy-biased models. Moreover, our method can also be applied to other bipartite graphs.

As an indispensable technique in the field of Information Filtering, Recommender System has been well studied and developed both in academia and in industry recently. However, most of current recommender systems suffer the following problems: (1) The large-scale and sparse data of the user-item matrix seriously affect the recommendation quality. As a result, most of the recommender systems cannot easily deal with users who have made very few ratings. (2) The traditional recommender systems assume that all the users are independent and identically distributed; this assumption ignores the connections among users, which is not consistent with the real world recommendations. Aiming at modeling recommender systems more accurately and realistically, we propose a novel probabilistic factor analysis framework, which naturally fuses the users' tastes and their trusted friends' favors together. In this framework, we coin the term Social Trust Ensemble to represent the formulation of the social trust restrictions on the recommender systems. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations, while the experimental results show that our method performs better than the state-of-the-art approaches.

Recently the re-ranking algorithms have been quite popular for web search and data mining. However, one of the issues is that those algorithms treat the content and link information individually. Inspired by graph-based machine learning algorithms, we propose a novel and general framework to model the re-ranking algorithm, by regularizing the smoothness of ranking scores over the graph, along with a regularizer on the initial ranking scores (which are obtained by the base ranker). The intuition behind the model is the global consistency over the graph: similar entities are likely to have the same ranking scores with respect to a query. Our approach simultaneously incorporates the content with other explicit or implicit link information in a latent space graph. Then an effective unified re-ranking algorithm is performed on the graph with respect to the query. To illustrate our methodology, we apply the framework to literature retrieval and expert finding applications on DBLP bibliography data. We compare the proposed method with the initial language model method and another PageRank-style re-ranking method. Also, we evaluate the proposed method with varying graphs and settings. Experimental results show that the improvement in our proposed method is consistent and promising.

Support vector machines (SVM) are state-of-the-art classifiers. Typically L2-norm or L1-norm is adopted as a regularization term in SVMs, while other norm-based SVMs, for example, the L0-norm SVM or even the L(infinity)-norm SVM, are rarely seen in the literature. The major reason is that L0-norm describes a discontinuous and nonconvex term, leading to a combinatorially NP-hard optimization problem. In this letter, motivated by Bayesian learning, we propose a novel framework that can implement arbitrary norm-based SVMs in polynomial time. One significant feature of this framework is that only a sequence of sequential minimal optimization problems needs to be solved, thus making it practical in many real applications. The proposed framework is important in the sense that Bayesian priors can be efficiently plugged into most learning methods without knowing the explicit form. Hence, this builds a connection between Bayesian learning and the kernel machines. We derive the theoretical framework, demonstrate how our approach works on the L0-norm SVM as a typical example, and perform a series of experiments to validate its advantages. Experimental results on nine benchmark data sets are very encouraging. The implemented L0-norm is competitive with or even better than the standard L2-norm SVM in terms of accuracy but with a reduced number of support vectors, -9.46% of the number on average. When compared with another sparse model, the relevance vector machine, our proposed algorithm also demonstrates better sparse properties with a training speed over seven times faster.

In this paper, we present a fusion approach to solve the nonrigid shape recovery problem, which takes advantage of both the appearance information and the local features. We have two major contributions. First, we propose a novel progressive finite Newton optimization scheme for the feature-based nonrigid surface detection problem, which is reduced to only solving a set of linear equations. The key is to formulate the nonrigid surface detection as an unconstrained quadratic optimization problem that has a closed-form solution for a given set of observations. Second, we propose a deformable Lucas-Kanade algorithm that triangulates the template image into small patches and constrains the deformation through the second-order derivatives of the mesh vertices. We formulate it into a sparse regularized least squares problem, which is able to reduce the computational cost and the memory requirement. The inverse compositional algorithm is applied to efficiently solve the optimization problem. We have conducted extensive experiments for performance evaluation on various environments, whose promising results show that the proposed algorithm is both efficient and effective.

Most machine learning tasks in data classification and information retrieval require manually labeled data examples in the training stage. The goal of active learning is to select the most informative examples for manual labeling in these learning tasks. Most of the previous studies in active learning have focused on selecting a single unlabeled example in each iteration. This could be inefficient, since the classification model has to be retrained for every acquired labeled example. It is also inappropriate for the setup of information retrieval tasks where the user's relevance feedback is often provided for the top K retrieved items. In this paper, we present a framework for batch mode active learning, which selects a number of informative examples for manual labeling in each iteration. The key feature of batch mode active learning is to reduce the redundancy among the selected examples such that each example provides unique information for model updating. To this end, we employ the Fisher information matrix as the measurement of model uncertainty, and choose the set of unlabeled examples that can efficiently reduce the Fisher information of the classification model. We apply our batch mode active learning framework to both text categorization and image retrieval. Promising results show that our algorithms are significantly more effective than the active learning approaches that select unlabeled examples based only on their informativeness for the classification model.

Support vector machine (SVM) active learning is one popular and successful technique for relevance feedback in content-based image retrieval (CBIR). Despite the success, conventional SVM active learning has two main drawbacks. First, the performance of SVM is usually limited by the number of labeled examples. It often suffers a poor performance for the small-sized labeled examples, which is the case in relevance feedback. Second, conventional approaches do not take into account the redundancy among examples, and could select multiple examples that are similar (or even identical). In this work, we propose a novel scheme for explicitly addressing the drawbacks. It first learns a kernel function from a mixture of labeled and unlabeled data, and therefore alleviates the problem of small-sized training data. The kernel will then be used for a batch mode active learning method to identify the most informative and diverse examples via a min-max framework. Two novel algorithms are proposed to solve the related combinatorial optimization: the first approach approximates the problem into a quadratic program, and the second solves the combinatorial optimization approximately by a greedy algorithm that exploits the merits of submodular functions. Extensive experiments with image retrieval using both natural photo images and medical images show that the proposed algorithms are significantly more effective than the state-of-the-art approaches. A demo is available at http://msm.cais.ntu.edu.sg/LSCBIR/.

Heat-diffusion models have been successfully applied to various domains such as classification and dimensionality-reduction tasks in manifold learning. One critical local approximation technique is employed to weigh the edges in the graph constructed from data points. This approximation technique is based on an implicit assumption that the data are distributed evenly. However, this assumption is not valid in most cases, so the approximation is not accurate in these cases. To solve this challenging problem, we propose a volume-based heat-diffusion model (VHDM). In VHDM, the volume is theoretically justified by handling the input data that are unevenly distributed on an unknown manifold. We also propose a novel volume-based heat-diffusion classifier (VHDC) based on VHDM. One of the advantages of VHDC is that the computational complexity is linear on the number of edges given a constructed graph. Moreover, we give an analysis on the stability of VHDC with respect to its three free parameters, and we demonstrate the connection between VHDC and some other classifiers. Experiments show that VHDC performs better than Parzen window approach, K nearest neighbor, and the HDC without volumes in prediction accuracy and outperforms some recently proposed transductive-learning algorithms. The enhanced performance of VHDC shows the validity of introducing the volume. The experiments also confirm the stability of VHDC with respect to its three free parameters.

The use of services, especially Web services, became a common practice. In Web services, standard communication protocols and simple broker-request architectures are needed to facilitate exchange of services, and this standardization simplifies interoperability. In the coming few years, services are expected to dominate software industry. There are increasing amount of Web services being made available in the Internet, and an efficient Web services composition algorithm would help to integrate different algorithm together to provide a variety of services. In this paper, we provide a dynamic Web service composition algorithm with verification of Petri-Nets. Each Web service is described by Web Service Definition Language (WSDL) and their interactions with other services are described by Web Service Choreography Interface (WSCI). Our algorithm compose the Web services with the information provided by these two descriptions. After the composition, we verify the Web service to be deadlock free with modeling the Web service as a Petri-Net. We conduct a series of experiments to evaluate the correctness and performance of the composed Web service.

For a given query raised by a specific user, the Query Suggestion technique aims to recommend relevant queries which potentially suit the information needs of that user. Due to the complexity of the Web structure and the ambiguity of users' inputs, most of the suggestion algorithms suffer from the problem of poor recommendation accuracy. In this paper, aiming at providing semantically relevant queries for users, we develop a novel, effective and efficient two-level query suggestion model by mining clickthrough data, in the form of two bipartite graphs (user-query and query-URL bipartite graphs) extracted from the clickthrough data. Based on this, we first propose a joint matrix factorization method which utilizes two bipartite graphs to learn the low-rank query latent feature space, and then build a query similarity graph based on the features. After that, we design an online ranking algorithm to propagate similarities on the query similarity graph, and finally recommend latent semantically relevant queries to users. Experimental analysis on the clickthrough data of a commercial search engine shows the effectiveness and the efficiency of our method.

Social Network Marketing techniques employ pre-existing social networks to increase brands or products awareness through word-of-mouth promotion. Full understanding of social network marketing and the potential candidates that can thus be marketed to certainly offer lucrative opportunities for prospective sellers. Due to the complexity of social networks, few models exist to interpret social network marketing realistically. We propose to model social network marketing using Heat Diffusion Processes. This paper presents three diffusion models, along with three algorithms for selecting the best individuals to receive marketing samples. These approaches have the following advantages to best illustrate the properties of real-world social networks: (1) We can plan a marketing strategy sequentially in time since we include a time factor in the simulation of product adoptions; (2) The algorithm of selecting marketing candidates best represents and utilizes the clustering property of real-world social networks; and (3) The model we construct can diffuse both positive and negative comments on products or brands in order to simulate the complicated communications within social networks. Our work represents a novel approach to the analysis of social network marketing, and is the first work to propose how to defend against negative comments within social networks. Complexity analysis shows our model is also scalable to very large social networks.

Data sparsity, scalability and prediction quality have been recognized as the three most crucial challenges that every collaborative filtering algorithm or recommender system confronts. Many existing approaches to recommender systems can neither handle very large datasets nor easily deal with users who have made very few ratings or even none at all. Moreover, traditional recommender systems assume that all the users are independent and identically distributed; this assumption ignores the social interactions or connections among users. In view of the exponential growth of information generated by online social networks, social network analysis is becoming important for many Web applications. Following the intuition that a person's social network will affect personal behaviors on the Web, this paper proposes a factor analysis approach based on probabilistic matrix factorization to solve the data sparsity and poor prediction accuracy problems by employing both users' social network information and rating records. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations, while the experimental results shows that our method performs much better than the state-of-the-art approaches, especially in the circumstance that users have made few or no ratings.

In automated text categorization, given a small number of labeled documents, it is very challenging, if not impossible, to build a reliable classifier that is able to achieve high classification accuracy. To address this problem, a novel web-assisted text categorization framework is proposed in this paper. Important keywords are first automatically identified from the available labeled documents to form the queries. Search engines are then utilized to retrieve from the Web a multitude of relevant documents, which are then exploited by a semi-supervised framework. To our best knowledge, this work is the first study of this kind. Extensive experimental study shows the encouraging results of the proposed text categorization framework: using Google as the web search engine, the proposed framework is able to reduce the classification error by 30% when compared with the state-of-the-art supervised text categorization method.

Active learning has been shown as a key technique for improving content-based image retrieval (CBIR) performance. Among various methods, support vector machine (SVM) active learning is popular for its application to relevance feedback in CBIR. However, the regular SVM active learning has two main drawbacks when used for relevance feedback. First, SVM often suffers from learning with a small number of labeled examples, which is the case in relevance feedback. Second, SVM active learning usually does not take into account the redundancy among examples, and therefore could select multiple examples in relevance feedback that are similar (or even identical) to each other. In this paper, we propose a novel scheme that exploits both semi-supervised kernel learning and batch mode active learning for relevance feedback in CBIR. In particular, a kernel function is first learned from a mixture of labeled and unlabeled examples. The kernel will then be used to effectively identify the informative and diverse examples for active learning via a min-max framework. An empirical study with relevance feedback of CBIR showed that the proposed scheme is significantly more effective than other state-of-the-art approaches.

It is critical to guarantee the reliability of service-oriented applications. This is because they may employ remote Web Services as components, which may easily become unavailable in the unpredictable Internet environment. This practical experience report presents a Distribute REliability Assessment Mechanism for Web Services (WS-DREAM), allowing users to carry out Web Services reliability assessment in a collaborative manner. With WS-DREAM, users in different geography locations help each other to carry out testing, and share test cases under the coordination of a centralized server. Based on this collaborative mechanism, reliability assessment for Web Services in real environment from different locations of the world becomes seamless. To illustrate the advantage of this mechanism, a prototype is implemented and a case study is carried out. Users from five locations all over the world perform reliability assessment to Web Services distributed in six countries. Over 1,000,000 test cases are executed in a collaborative manner and detailed results are provided.

In this paper, we propose a novel coalitional game model for security issues in wireless networks. The model can be applied to not only mobile ad hoc networks but also wireless sensor networks. We define a new throughput characteristic function, on the basis of which nodes are enforced to cooperate and form coalitions. This function implies the maximal throughput and the most reliable traffic that a coalition can achieve. The fair payoff share inside the coalition is given by Shapley value after proving the feasibility of this method. Then a set of game rules is presented to establish a threatening mechanism to all players. We then describe the coalition formation procedure and explain how to integrate this game theoretic model with available wireless routing protocols. Finally, a theoretical analysis is conducted to illustrate the convergence situation and verify the correctness of the formulation.

This paper discusses a point-distribution index, l, which measures the normalized minimum distance between sensors. Maximizing l of a set of points causes the Delaunay triangulation graph of these points to be a net of equilateral triangles. Such a structure indicates the lowest redundancy of coverage if each point represents the center of a disc. Thus l can serve as a promising measure for solving a critical problem in field coverage: How to group a set of sensor nodes into disjoint subsets so that each subset can cover the entire field? Based on the l index, we develop an effective algorithm, MAXINE (MAXimizing-l Node-redundancy Exploiting), for the sensor- grouping problem. We evaluate the performance of MAXINE through extensive simulations and compare it with existing algorithms. The results demonstrate the effectiveness of MAXINE and verify the superiority of employing i for the sensor-grouping problem.

Finding relevant experts in a specific field is often crucial for consulting, both in industry and in academia. The aim of this paper is to address the expert-finding task in a real world academic field. We present three models for expert finding based on the large-scale DBLP bibliography and Google scholar for data supplementation. The first, a novel weighted language model, models an expert candidate based on the relevance and importance of associated documents by introducing a document prior probability, and achieves much better results than the basic language model. The second, a topic-based model, represents each candidate as a weighted sum of multiple topics, whilst the third, a hybrid model, combines the language model and the topic-based model. We evaluate our system using a benchmark dataset based on human relevance judgments of how well the expertise of proposed experts matches a query topic. Evaluation results show that our hybrid model outperforms other models in nearly all metrics.

Zero-norm, defined as the number of non-zero elements in a vector, is an ideal quantity for feature selection. However, minimization of zero-norm is generally regarded as a combinatorially difficult optimization problem. In contrast to previous methods that usually optimize a surrogate of zero-norm, we propose a direct optimization method to achieve zero-norm for feature selection in this paper. Based on Expectation Maximization (EM), this method boils down to solving a sequence of Quadratic Programming problems and hence can be practically optimized in polynomial time. We show that the proposed optimization technique has a nice Bayesian interpretation and converges to the true zero norm asymptotically, provided that a good starting point is given. Following the scheme of our proposed zero-norm, we even show that an arbitrary-norm based Support Vector Machine can be achieved in polynomial time. A series of experiments demonstrate that our proposed EM based zero-norm outperforms other state-of-the-art methods for feature selection on biological microarray data and UCI data, in terms of both the accuracy and the learning efficiency.

We consider the problem of semi-supervised learning (SSL) from general unlabeled data, which may contain irrelevant samples. Within the binary setting, our model manages to better utilize the information from unlabeled data by formulating them as a three-class (-1,+1, 0) mixture, where class 0 represents the irrelevant data. This distinguishes our work from the traditional SSL problem where unlabeled data are assumed to contain relevant samples only, either +1 or -1, which are forced to be the same as the given labeled samples. This work is also different from another family of popular models, universum learning (universum means "irrelevant" data), in that the universum need not to be specified beforehand. One significant contribution of our proposed framework is that such irrelevant samples can be automatically detected from the available unlabeled data, even though they are mixed with relevant data. This hence presents a general SSL framework that does not force "clean" unlabeled data.More importantly, we formulate this general learning framework as a Semi-definite Programming problem, making it solvable in polynomial time. A series of experiments demonstrate that the proposed framework can outperform the traditional SSL on both synthetic and real data.

Redundancy-based fault tolerance strategies are proposed for building reliable Service-Oriented Architectures/Applications (SOA), which are usually developed on the unpredictable remote Web services. This paper proposes and implements a distributed replication strategy evaluation and selection framework for fault tolerant Web services. Based on this framework, we provide a systematic comparison of various replication strategies by theoretical formula and real-world experiments. Moreover, a user participated strategy selection algorithm is designed and verified. Experiments are conducted to illustrate the advantage of this framework. In these experiments, users from six different locations all over the world perform evaluation of Web services distributed in six countries. Over 1,000,000 test cases are executed in a collaborative manner and detailed results are also provided.

Minimax Probability Machine (MPM), learning a decision function by minimizing the maximum probability of misclassification, has demonstrated very promising performance in classification and regression. However, MPM is often challenged for its slow training and test procedures. Aiming to solve this problem, we propose an efficient model named Minimax Clustering Probability Machine (MCPM). Following many traditional methods, we represent training data points by several clusters. Different from these methods, a Generalized Probability Product Kernel is appropriately defined to grasp the inner distributional information over the clusters. Incorporating clustering information via a non-linear kernel, MCPM can fast train and test in classification problem with promising performance. Another appealing property of the proposed approach is that MCPM can still derive an explicit worst-case accuracy bound for the decision boundary. Experimental results on synthetic and real data validate the effectiveness of MCPM for classification while attaining high accuracy.

Recent experience and feedback from panels indicates that what the audience likes best is the chance to ask questions, particularly regarding things that might solve problems on their development projects or in their research studies. This panel has been held at every ISSRE since 1997, and has been highly popular. Originally the panel was chaired by John Musa, but Mike Hinchey took over that role last year. The panel has no presentations, only questions. The questions can be on anything, ranging from theory to details of application. The panelists have been selected to make available wide and extensive experience in the field.

Reliability is a key issue of the service-oriented architecture (SOA) which is widely employed in critical domains such as e-commerce and e-government. Redundancy-based fault tolerance strategies are usually employed for building reliable SOA on top of unreliable remote Web services. Based on the idea of user-collaboration, this paper proposes a QoS-aware middleware for fault tolerant Web services. Based on this middleware, service-oriented applications can dynamically adjust their optimal fault tolerance strategy to achieve good service reliability as well as good overall performance. A dynamic fault tolerance replication strategy is designed and evaluated. Experiments are conducted to illustrate the advantage of the proposed middleware as well as the dynamic fault tolerance replication strategy. Comparison of the effectiveness of the proposed dynamic fault tolerance strategy and various traditional fault tolerance strategies are also provided.

Near-duplicate image retrieval plays an important role in many real-world multimedia applications. Most previous approaches have some limitations. For example, conventional appearance-based methods may suffer from the illumination variations and occlusion issue, and local feature correspondence-based methods often do not consider local deformations and the spatial coherence between two point sets. In this paper, we propose a novel and effective Nonrigid Image Matching (NIM) approach to tackle the task of near-duplicate keyframe retrieval from real-world video corpora. In contrast to previous approaches, the NIM technique can recover an explicit mapping between two nearduplicate images with a few deformation parameters and find out the correct correspondences from noisy data effectively. To make our technique applicable to large-scale applications, we suggest an effective multi-level ranking scheme that filters out the irrelevant results in a coarse-to-fine manner. In our ranking scheme, to overcome the extremely small training size challenge, we employ a semi-supervised learning method for improving the performance using unlabeled data. To evaluate the effectiveness of our solution, we have conducted extensive experiments on two benchmark testbeds extracted from the TRECVID2003 and TRECVID2004 corpora. The promising results show that our proposed method is more effective than other state-of-the-art approaches for near-duplicate keyframe retrieval.

We consider the problem of multiple kernel learning (MKL), which can be formulated as a convex-concave problem. In the past, two efficient methods, i.e., Semi-Infinite Linear Programming (SILP) and Subgradient Descent (SD), have been proposed for large-scale multiple kernel learning. Despite their success, both methods have their own shortcomings: (a) the SD method utilizes the gradient of only the current solution, and (b) the SILP method does not regularize the approximate solution obtained from the cutting plane model. In this work, we extend the level method, which was originally designed for optimizing non-smooth objective functions, to convex-concave optimization, and apply it to multiple kernel learning. The extended level method overcomes the drawbacks of SILP and SD by exploiting all the gradients computed in past iterations and by regularizing the solution via a projection to a level set. Empirical study with eight UCI datasets shows that the extended level method can significantly improve efficiency by saving on average 91.9% of computational time over the SILP method and 70.3% over the SD method.

Regularized Least Squares (RLS) algorithms have the ability to avoid over-fitting problems and to express solutions as kernel expansions. However, we observe that the current RLS algorithms cannot provide a satisfactory interpretation even on a constant function. On the other hand, while kernel-based algorithms have been developed in such a tendency that almost all learning algorithms are kernelized or being kernelized, a basic fact is often ignored: The learned function from the data and the kernel fits the data well, but may not be consistent with the kernel. Based on these considerations and on the intuition that a good kernel-based inductive function should be consistent with both the data and the kernel, a novel learning scheme is proposed. The advantages of this scheme lie in its corresponding Representer Theorem, its strong interpretation ability about what kind of functions should not be penalized, and its promising accuracy improvements shown in a number of experiments. Furthermore, we provide a detailed technical description about heat kernels, which serves as an example for the readers to apply similar techniques for other kernels. Our work provides a preliminary step in a new direction to explore the varying consistency between inductive functions and kernels under various distributions.

Many state-of-the-art wireless sensor networks have been equipped with reprogramming modules, e.g., those for software/firmware updates. System migration tasks such as software reprogramming however will interrupt normal sensing and data reporting operations of a sensor node. Although such tasks are

A critical issue of large-scale multimedia retrieval is how to develop an effective framework for ranking the search results. This problem is particularly challenging for content-based video retrieval due to some issues such as short text queries, insufficient sample learning, fusion of multimodal contents, and large-scale learning with huge media data. In this paper, we propose a novel multimodal and multilevel (MMML) ranking framework to attack the challenging ranking problem of content-based video retrieval. We represent the video retrieval task by graphs and suggest a graph based semi-supervised ranking (SSR) scheme, which can learn with small samples effectively and integrate multimodal resources for ranking smoothly. To make the semi-supervised ranking solution practical for large-scale retrieval tasks, we propose a multilevel ranking framework that unifies several different ranking approaches in a cascade fashion. We have conducted empirical evaluations of our proposed solution for automatic search tasks on the benchmark testbed of TRECVID2005. The promising empirical results show that our ranking solutions are effective and very competitive with the state-of-the-art solutions in the TRECVID evaluations.

Face annotation in images and videos enjoys many potential applications in multimedia information retrieval. Face annotation usually requires many training data labeled by hand in order to build effective classifiers. This is particularly challenging when annotating faces on large-scale collections of media data, in which huge labeling efforts would be very expensive. As a result, traditional supervised face annotation methods often suffer from insufficient training data. To attack this challenge, in this paper, we propose a novel Transductive Kernel Fisher Discriminant (TKFD) scheme for face annotation, which outperforms traditional supervised annotation methods with few training data. The main idea of our approach is to solve the Fisher's discriminant using deformed kernels incorporating the information of both labeled and unlabeled data. To evaluate the effectiveness of our method, we have conducted extensive experiments on three types of multimedia testbeds: the FRGC benchmark face dataset, the Yahoo! web image collection, and the TRECVID video data collection. The experimental results show that our TKFD algorithm is more effective than traditional supervised approaches, especially when there are very few training data.

In this paper, we propose a novel large margin classifier, called the maxi-min margin machine M(4). This model learns the decision boundary both locally and globally. In comparison, other large margin classifiers construct separating hyperplanes only either locally or globally. For example, a state-of-the-art large margin classifier, the support vector machine (SVM), considers data only locally, while another significant model, the minimax probability machine (MPM), focuses on building the decision hyperplane exclusively based on the global information. As a major contribution, we show that SVM yields the same solution as M(4) when data satisfy certain conditions, and MPM can be regarded as a relaxation model of M(4). Moreover, based on our proposed local and global view of data, another popular model, the linear discriminant analysis, can easily be interpreted and extended as well. We describe the M(4) model definition, provide a geometrical interpretation, present theoretical justifications, and propose a practical sequential conic programming method to solve the optimization problem. We also show how to exploit Mercer kernels to extend M(4) for nonlinear classifications. Furthermore, we perform a series of evaluations on both synthetic data sets and real-world benchmark data sets. Comparison with SVM and MPM demonstrates the advantages of our new model.

Robust regression techniques are critical to fitting data with noise in real-world applications. Most previous work of robust kernel regression is usually formulated into a dual form, which is then solved by some quadratic program solver consequently. In this correspondence, we propose a new formulation for robust regularized kernel regression under the theoretical framework of regularization networks and then tackle the optimization problem directly in the primal. We show that the primal and dual approaches are equivalent to achieving similar regression performance, but the primal formulation is more efficient and easier to be implemented than the dual one. Different from previous work, our approach also optimizes the bias term. In addition, we show that the proposed solution can be easily extended to other noise-reliable loss function, including the Huber- epsilon insensitive loss function. Finally, we conduct a set of experiments on both artificial and real data sets, in which promising results show that the proposed method is effective and more efficient than traditional approaches.