As software systems scale, fault detection and localization become increasingly complex due to intricate module interactions. Logging is essential for diagnosis, yet an analysis of 1158 bug reports from the Bugs.jar dataset shows that faultrelated variables have only 13.16% log coverage, leading to incomplete diagnostics, prolonged troubleshooting, and higher maintenance costs. Existing logging enhancement methods focus on source code analysis during development, lacking adaptability to deployed systems, while runtime modifications are often costly and impractical. To address these challenges, this paper presents VarFR, a novel bytecode-level approach for dynamically tracking fault-related variables and enhancing log coverage without modifying the source code. VarFR employs a recommendation model that constructs a bytecode multi-feature fusion graph by integrating bytecode semantics, method control flow, and local variable metadata. By improving fault observability at the bytecode level, the proposed approach has the potential to facilitate debugging, reduce maintenance overhead, and enhance software adaptability, thereby supporting the long-term evolution of software systems. Experimental results on the Bugs.jar dataset demonstrate that VarFR significantly outperforms baseline models in fault-related variable recommendation, underscoring its effectiveness in improving software maintainability and reliability.

Service-oriented architectures involve complex interactions, making failure detection and diagnosis challenging. Traditional static log statements often miss critical variables and lack runtime flexibility, resulting in limited observability. To address this, we propose a bytecode-level variable monitoring approach using the ASM library. Our tool transparently instruments service bytecode, enabling dynamic, source-free monitoring with minimal overhead. We further introduce a fusion model that analyzes bytecode semantics and structure to recommend critical variables at runtime. This enhances service observability and supports more efficient failure diagnosis.

The task of converting monolithic programs to mi-croservices architecture is complex, hindered by the intertwined nature of program control and data flows. Conventional methods for microservice extraction often fall short in capturing the essen-tial connections within a monolithic structure and in propagating properties to distant neighbors for effective clustering. To address these issues, we introduce an innovative graph-based deep clustering technique that utilizes both control flow and data flow graphs. This approach offers a thorough analysis of class interactions within monolithic applications, facilitating accurate identification and extraction of microservices. Furthermore, we present the Microservice Extraction Graph Neural Network (MEGNN), an advanced graph attention network designed to enhance message transmission depth and enable nodes to assimilate features from k-hop neighbors. This method extends the reach of message distribution across node chains and mitigates the issue of feature homogenization, leading to more cohesive clustering of related nodes and improving the quality of microservices extraction. Experimental evaluations on data from three publicly accessible Java monolithic programs confirm that our proposed method surpasses existing techniques in microservices extraction efficacy.

Microservices architecture, known for its advantages in scalability and availability, presents challenges in pinpointing failure causes due to its inherent complexity. Traditional root cause analysis (RCA) methods often lack comprehensive insights into anomaly detection and exhibit shortcomings in finegrained anomaly localization. To address these gaps, we propose RootScan, an innovative framework tailored for fine-grained and interpretable root cause analysis in microservices. Our approach features an event-based observability model that unifies multimodal data (traces, metrics, and logs) into a structured event flow, providing a holistic view of system behavior. Then, we leverage the interpretability of generative networks to detect anomalies from the event flow level to feature level, yielding valuable clues for subsequent root cause analysis. Further, we explore anomaly propagation patterns to precisely trace root causes at fine-grained levels (container, microservice and component). We validate our proposal using an open-source dataset and benchmark microservice environment. Experimental results demonstrate the superiority of our approach, markedly enhancing RCA accuracy and reliability in microservices environments.

Convolutional Neural Network (CNN) and Transformer architectures have been extensively applied in the domain of remote sensing image super-resolution. However, to achieve optimal performance, many existing methods are designed with a large number of parameters, thereby increasing the complexity of the model and hindering practical deployment. To address this issue, we propose an innovative model that synergizes CNN and Transformer architectures while employing a minimal number of their respective modules. This approach significantly reduces the parameter count while maintaining superior performance. Firstly, by constructing additional shallow feature representations as input, we enhance the feature extraction capabilities for individual images. Secondly, we utilize residual connections between various modules to integrate multi-scale, high-dimensional feature information, thus ensuring efficient transmission. Finally, the image reconstruction module is employed to restore the high-resolution image. Experimental results show that FCTNet significantly outperforms existing methods while maintaining a substantially lower parameter count, as demonstrated through evaluations on two public datasets.

The increasing software vulnerabilities underscore the urgency of effective vulnerability detection for safeguarding security and socio-economic stability. Existing pre-trained model-based methods for vulnerability detection often suffer from limitations such as the compression of high-dimensional features and the omission of valuable semantic label information, which collectively compromise performance metrics and fine-tuning efficacy. To address these issues, we propose a novel methodology for fine-tuning pre-trained models using prompt learning. By projecting classification labels as prompts into a high-dimensional feature space, our model preserves essential spatial structures and optimizes the fine-tuning process more effectively. To further enhance classification performance, we design a specialized prompt network for learning adaptable prompts, along with a code encoder network that synergistically captures the semantic and structural nuances of code. Extensive empirical evaluation on two publicly available datasets, SARD and CodeXGLUE demonstrates significant improvements in classification performance and prompt optimization compared to existing state-of-the-art models. Specifically, our method achieves an accuracy of 87.98% on the SARD dataset and 70.97% on the CodeXGLUE dataset, marking a remarkable improvement over state-of-the-art solutions.

Issue reports in software repositories serve as a vital channel for users to report problems, bugs, and provide valuable suggestions for project improvements. Issues are typically assigned a diverse range of labels, enabling categorization and organization. However, the manual labeling of these reports becomes laborious and challenging due to the sheer volume of projects and issue reports. Existing automatic labeling solutions often rely on text matching or multi-classification approaches, which may overlook label dependencies and contextual correlations, compromising the accuracy and relevance of issue classification. To address this challenge, we propose an innovative framework named GlobalTagNet that enhances the automation of issue label assignment. Our approach leverages the power of Heterogeneous Graph Transformer (HGT) to explore label dependencies and semantics. By utilizing hierarchical graph structures, our model captures global label dependencies and correlations, enabling effective learning of label relationships from a holistic perspective. Furthermore, our approach enables the integration of information across multiple levels of the graph, facilitating more accurate inference of label relationships. We conducted comprehensive experiments to assess the effectiveness of our proposed approach. The experimental results unequivocally demonstrate that our method surpasses the performance of baseline solutions, achieving a remarkable improvement of up to 6% in Accuracy, and 8% in F1-score.

In today’s urban transportation landscape, the accurate and timely prediction of traffic conditions holds immense significance for optimizing traffic flow and resource allocation. However, conventional centralized methods for traffic prediction come with the drawback of data collection from various roads, a process that can introduce latency and privacy concerns. To address these pressing challenges, our innovative approach seamlessly integrates road spatiotemporal data, edge computing, and privacy protection inspired by federated learning. We utilize Long Short-Term Memory (LSTM) networks to make real-time predictions of vehicle speeds on individual road segments. Furthermore, we introduce Graph Neural Networks (GNN) to effectively amalgamate predictive insights from the target road and its neighboring roads, effectively capturing intricate spatial relationships within traffic data. Our approach excels in reducing data requirements and alleviating computational burdens, enabling straightforward deployment on edge servers tailored to individual road segments. Most importantly, it prioritizes privacy by circumventing the exchange of raw data, instead opting for the selective sharing of predictions between segments. Our rigorous experimental results demonstrate that our approach outperforms traditional federated averaging algorithms, underscoring its remarkable security attributes while achieving superior predictive accuracy.

API recommendation has been widely used to enhance developers’ efficiency in software development. However, existing API recommendation methods for dynamic languages such as Python usually suffer from the limitations of incorrect type inference and lack of rich contextual semantics. To address these issues, we propose in this paper a novel approach, PyBartRec, to recommend APIs for Python programs concerning their rich semantics. Instead of analyzing the data flow information only, our approach utilizes a Transformer-based pre-trained model to extract the semantic features of Python code snippets. Such contextual information allows our approach to recommend correct APIs even when the APIs are not included in the local data flow information. We also use such information to perform a post-processing type inference. By narrowing the range of candidate types, our approach can recommend APIs accurately even in the failure scenarios of type inference. We evaluated PyBartRec in eight popular Python projects and the experimental results show that our approach significantly outperforms the state-of-the-art solutions. In particular, the average top-1 accuracy and average top-10 accuracy of PyBartRec within the same project are over 40%, and 60%, respectively. In cross-project recommendation, the MRR of PyBartRec is also over 40%.

Deep learning Transformer architectures play a critical role in developing advanced code prediction models, which are essential in modern Integrated Development Environments (IDEs). Nevertheless, these architectures encounter a significant challenge in effectively capturing and utilizing the structural information present in Abstract Syntax Trees (ASTs). To tackle this challenge, we propose an innovative approach that leverages AST structural matrices to enhance Transformers for source code prediction. Specifically, we integrate three types of AST structural matrices - the R matrix, A&S matrix, and MVG matrix - into the attention module of the Transformer to effectively capture the structural information within ASTs. To ensure optimal utilization, we have devised a range of strategies and integration methods tailored specifically for the attention module. To assess the effectiveness of our proposal, we conduct empirical studies using a standard Python dataset. The results demonstrate that incorporating AST structural matrices significantly enhances the accuracy of code prediction models, leading to an overall improvement from 73.18% to 75.10%. Furthermore, we conduct an in-depth analysis of the impact of each matrix type, offering a comprehensive understanding of their application scenarios. This insightful analysis provides valuable guidance on how to effectively leverage each matrix type in various contexts.

The code clone detection method based on semantic similarity has important value in software engineering tasks (e.g., software evolution, software reuse). Traditional code clone detection technologies pay more attention to the similarity of code at the syntax level, and less attention to the semantic similarity of the code. As a result, candidate codes similar in semantics are ignored. To address this issue, we propose a code clone detection method based on semantic similarity. By treating code as a series of interdependent events that occur continuously, we design a model namely EDAM to encode code semantic information based on event embedding and event dependency. The EDAM model uses the event embedding method to model the execution characteristics of program statements and the data dependence information between all statements. In this way, we can embed the program semantic information into a vector and use the vector to detect codes similar in semantics. Experimental results show that the performance of our EDAM model is superior to state-of-the-art open source models for code clone detection.

Matching APP reviews with bug reports can help APP developers to quickly identify new bugs from the users’ feedback. Existing solutions represent the semantics of APP reviews and bug reports via carefully designed features and models, the performance of which however depends heavily on the manually designed model and the training data set. Large-scale pretrained models can well capture the semantics of text and have demonstrated their success in many NLP tasks. Inspired by this, we explore the effect of various pretrained models on the matching accuracy of app review and bug report. We conduct a systematic study to analyze the factors of four major pretrained models (including T5, Sentence T5, Sentence MiniLM, Sentence BERT and so on) on the matching accuracy. We find that the accuracy of Sentence T5 and Sentence MiniLM in four open source applications is significantly greater than that of the state-of-the-art approach DeepMatcher. Based on the findings, we design a novel approach to match the APP reviews with bug reports based on the pretrained model Sentence T5 and Sentence MiniLM to calculate the sentence similarity. We test it on four open source applications and the results show that our method outperforms the existing solution. On average, the precision of Sentence T5 and Sentence MiniLM are increased by 17% and 13%, respectively, and the hit ratio are increased by 15% and 14%, respectively.

Mobile application markets usually enact policies to describe in detail the minimum requirements that an application should comply with. User comments on mobile applications contain a large amount of information that can be used to find out APP's violations of market policies in a cost-effective way. Existing state-of-the-art methods match user comments with the violations of market policies based on well-designed syntax rules, which however cannot well capture the semantics of user comments and cannot be generalized to the scenarios not covered by the rules. To address this issue, we propose an innovative method, UBC-BERT, to detect undesired behavior from user comments based on their semantics. By incorporating sentence embeddings with attention, we train a classification model for 21 groups of undesirable behaviors based on the fine-tuning of a pre-trained model BERT-BASE. The experimental results show that our solution outperforms the baseline solutions in terms of a higher precision(up to 60.5% more).

Distributed cache is one of the most important components in cloud computing and microservice systems. Adding cache components to the microservice system can significantly improve the concurrency and throughput of the whole microservice framework, and effectively promote the overall performance of the microservice system. However, traditional caching algorithm can easily lead to the imbalance of cache resource allocation in microservice system, thus affecting the performance of the microservice system. We study the basic architecture of cache in microservice platform, and propose the mathematical model of distributed cache and the probability distribution model of data access in microservice platform. Next, based on the above mathematical model, we design a cache replacement algorithm based on dynamic constraints in microservice platform. The algorithm can effectively solve a series of problems of traditional algorithms in microservice platform, making the algorithm more suitable for the environment of microservice architecture. We evaluate and verify the algorithm on the simulation software platform. Experimental results show that the algorithm meets the performance requirements of high cache hit rate, high concurrency and high throughput in the microservice platform, and is significantly better than the traditional cache algorithms.

Sentiment analysis plays an indispensable role to help understand people’s opinions automatically based on their reviews. Existing research on sentiment analysis mainly focuses on film reviews, e-commerce reviews and other fields. These work cannot be applied to analyze the sentiment of travel reviews directly because the mainstream commodity review dataset is richer and more regular than that of travel review dataset. More specifically, the special characteristic of travel reviews makes existing solutions fail to achieve satisfactory results. To address this issue, we first construct a travel review data set for sentiment analysis. Then, we conduct a systematic study to investigate and compare the factors that may affect the accuracy of sentiment analysis for travel reviews. Based on the study findings, we design a lightweight Glove-BiLSTM-CNN model and BERT-BiLSTM-CNN to analyze the sentiment for travel reviews. Experimental results show that our proposed models outperform the baseline solutions.

The rapid development of tourism economy has brought new challenges such as the prevention of consumer fraud for smart city applications. Traditional fraud detection approaches such as telecommunication fraud and credit card fraud detection need a data set containing both the normal behavior and abnormal behavior. Therefore, they are incompetent to address such challenges in the tourism market where the data set is open and very few records of fraud behaviors are available. To address this issue, we propose a P-feature conversion algorithm to construct third-party features to expose the outliers. These features reveal the internal characteristics of different businesses and their useful internal connections. Then, we build a fraud detection model for tourism based on the Local Outlier Factor anomaly detection algorithm. Experimental results show that our model can effectively identify fraudulent merchants in the tourism market.

Different news events have different effects on stock price changes. If they are simply fed to the neural network for prediction, the accuracy will be affected. We propose a method to predict stock price trend based on time series news information. First, we extract events from news text and represent them as dense vectors by event embedding technique. Further-more, we employ attention mechanism to figure out event is the main cause of the price fluctuation. Then, we use a Gated Recurrent Unit to model the influence of events on stock market. Experimental results show that our model achieve a certain improvement on S&P500 index compared to baseline methods.

Stock trend forecasting plays a great role in maximizing the profit of stock investment. However, due to the high volatility and non-stationarity of the stock market, accurate trend prediction is very difficult. With the development of the Internet and deep learning technology, people can use deep learning methods to reveal market trends and volatility from the explosive information on the Internet. Unfortunately, there is a large amount of content related to the stock market, and a large part of it is useless information. As a result, how to extract the effective information and combine this information as different characteristics to effectively predict stock trends has become the biggest challenge. In order to cope with these challenges, we use TextCNN as the news text feature extractor for feature extraction of news information, and propose a prediction method based on multi-feature fusion: Bi-LSTNAA, to predict the Chinese stock market. Extensive experiments on actual stock market data show that the our method has a greater improvement in the accuracy of stock trend prediction.

Sentiment analysis (SA) has been applied to various fields of software engineering (SE), such as app reviews, stack overflow Q&A website and API comments. General SA tools are trained based on movie or product review data. Research has shown that these SA tools can produce negative results when applied to the field of SE. In order to overcome the above limitations, developers need to customize tools (e.g., SentiStrength-SE, SentiCR, Senti4SD). In recent years, the pre-trained transformer-based models have brought great breakthroughs in the field of natural language processing. Therefore, we intend to fine-tune the pre-trained model BERT for downstream text classification tasks. We compare the performance of SE-specific tools. Meanwhile, we also studied the performance of SE-specific tools in a cross-platform setting. Experimental results show that our approach (BERT-FT) outperforms the existing state-of-the-art models in terms of F1-scores.

Semantic code clone detection is an important yet challenging task in software engineering. Traditional methods rely on expert experience and cannot automatically determine which features are better for semantic code clone detection. Moreover, the program dynamics (e.g., the execution characteristics and execution order of statements) are not considered in these methods. As a result, this limits their ability to detect semantic clones. To address this issue, we propose a code clone detection method based on event embedding tree and Graph Attention Network. Our method uses a program control flow graph to capture the execution characteristics of each statement and extract the context relationship of different statements in the control flow. Based on such information, our method can calculate the functional similarity of two pieces of code, thereby identifying semantically similar code fragments. Experimental results show that our method is superior to state-of-the-art open source methods for Type-3 (syntactic) / Type-4 (semantic) clone detection.

The development of intelligent transportation systems usually needs to predict the traffic conditions under a large data volume. Existing approaches usually use a single source of data and the impacts of the neighborhood road sections are not concerned. As a result, their prediction accuracy is usually compromised. To address this issue, we propose a recurrent neural network to predict the road conditions simultaneously concerning the information of multiple road sections at the same time. By perceiving the connectivity between multiple road sections and capturing their mutual influence, our model can significantly improve the prediction accuracy. The experiments based on two real-life dataset shows that our model outperforms the baseline model.

Web service is a popular solution to integrate components when building a software system, or to allow communication between a system and third-party users, providing a flexible and reusable mechanism to access its functionalities. Various web service based systems are prevailing in health service provision. We propose a framework of construction, searching and protection of typed health resources in terms of data, information and knowledge through a hierarchy consisting of Data Graph, Information Graph and Knowledge Graph. We use cases to illustrate the mechanism of the framework.

We identify that to improve the reusability of implementation in service engineering, we need to collect and modulate reusable strategy, knowledge and experience crossing business modeling, knowledge management and economic analysis simultaneously from an interdisciplinary perspective. Strategically to ease the complexity, we adopt the existing experience from design patterns by forming a service design pattern called service value broker(SVB). SVB can efficiently integrate the three domains with relieved complexity, enhanced reusability and efficiency, etc, catering different abstraction levels. In this paper, we focus on modeling the foundational aspects of SVB presentation and value transaction, etc.

E-Service related technologies have a profound and revolutionary impact on many traditional areas. We try to outline its influence on economics which we call E-service Economics at fundamental level. We model the new characteristics of this trend along two themes: value driven decision making which covers both reuse based value added analysis and service value broker (SVB) pattern which embodies the integration of information technology and business wisdoms, and E-contract based realization which equally supports collaborative and competitive business activities. An initial case based on E-contract to show the attained flexibility, dynamic and consistency of automation is demonstrated.

The service value broker(SVB) pattern integrates business modeling, knowledge management and economic analysis with relieved complexity, enhanced reusability and efficiency, etc. The study of SVB is an emerging interdisciplinary subject which will help to promote the reuse of knowledge, strategy and experience in service based designs and solutions. In this paper, we focus on enumerating collected SVBs empirically with initial analysis on their composition manners. The results from this paper will play a dominating role in fueling a coming E-service Economics era.