Universal graph pre-training has emerged as a key paradigm in graph representation learning, offering a promising way to train encoders to learn transferable representations from unlabeled graphs and to effectively generalize across a wide range of downstream tasks. However, recent explorations in universal graph pre-training primarily focus on homogeneous graphs and it remains unexplored for heterogeneous graphs, which exhibit greater structural and semantic complexity. This heterogeneity makes it highly challenging to train a universal encoder for diverse heterogeneous graphs: (i) the diverse types with dataset-specific semantics hinder the construction of a unified representation space; (ii) the number and semantics of meta-paths vary across datasets, making encoding and aggregation patterns learned from one dataset difficult to apply to others. To address these challenges, we propose a novel Meta-path-aware Universal heterogeneous Graph pre-training (MUG) approach. Specifically, for challenge (i), MUG introduces a input unification module that integrates information from multiple node and relation types within each heterogeneous graph into a unified representation. This representation is then projected into a shared space by a dimension-aware encoder, enabling alignment across graphs with diverse schemas. Furthermore, for challenge (ii), MUG trains a shared encoder to capture consistent structural patterns across diverse meta-path views rather than relying on dataset-specific aggregation strategies, while a global objective encourages discriminability and reduces dataset-specific biases. Extensive experiments demonstrate the effectiveness of MUG on some real datasets.

Document Visual Question Answering (DocVQA) aims to generate answers by jointly understanding the textual, layout, and visual elements within document images.Although end-to-end vision-based generative methods have reduced dependency on OCR, they still struggle to achieve precise evidence localization when page semantics are complex and highly similar.However, existing research lacks an in-depth theoretical analysis of the question-driven semantic representation space, failing to fundamentally address the distinguishability problem among semantically similar pages.To fill this theoretical gap, we propose and prove that, given a specific question, each page possesses a unique semantic representation, and there exists a bijective mapping between the page and its unique semantics.Based on this theoretical foundation, we introduce the Flow-Based Page Unique Semantic Mapping Architecture (FUMA), which reconstructs evidence localization from similarity-based retrieval into precise selection on unique semantics.FUMA employs fine-grained cross-modal attention to extract discriminative cues and utilizes flow-based reversible transformations with likelihood regularization to learn bijective mappings, ensuring that each page obtains a unique semantic representation.Moreover, a multi-expert collaboration mechanism complementarily models fine-grained multimodal information within each page, achieving robust answer generation.Experimental results demonstrate that FUMA significantly outperforms existing methods in both evidence localization and answer generation.

The evaluation of Kleene Closure Property Path Queries (KPPQs) over RDF graphs presents a fundamental trade-off. Under the intuitive simple path semantics, which forbids node repetition, the problem is NP-complete. This forces practical systems to adopt existential path semantics, which admits tractable evaluation but produces result sets bloated with redundant and often semantically meaningless paths due to unnecessary detours through cycles. This paper introduces a novel indexing framework that effectively navigates this trade-off. Our key insight is that while enumerating all simple paths is intractable, a carefully selected, compact set of representative simple paths is sufficient to correctly and completely answer connectivity queries in practice. We propose the Node Clustered Index (NCI), which pre-materializes such paths—including the longest acyclic chains and fundamental cycles —transforming online query processing from a costly graph traversal into an efficient index lookup and expansion operation. Crucially, our method guarantees that every result conforms to simple path semantics. Extensive experiments demonstrate that our approach achieves near-perfect recall, outperforming state-of-the-art systems by up to an order of magnitude. This work establishes that for NP-complete path problems, seeking polynomial-time, empirically perfect approximations is a viable and powerful paradigm, delivering both performance and quality without compromise.

Recently, Graph Foundation Models (GFMs) have emerged as a central focus in the field of graph learning due to their strong generalizability to various unseen graphs. However, existing GFMs typically work under the homophily assumption, and the exploration of universality on heterophilic graphs is still in its early stages. In fact, even in homophilic graphs, there exists limited yet informative heterophilic information that is not fully exploited by current GFMs. Moreover, due to the requirement for universality, the heterophily issue faced by GFMs is more challenging than in classical graph learning, as it requires training a single model to adapt to varying structures, features, and tasks. Classic heterophilic graph learning methods primarily based on the node-level homophily or heterophily. However, we highlight that homophily and heterophily exist not only at the node semantic level, but also at a finer granularity across individual feature dimensions. This finding enables GFMs to adapt to heterophilic graphs and better utilize the small amount of heterophilic information in homophilic graphs. Based on this, we propose Topology-aware Feature Sorting Graph Foundation Model (TFSGFM), which employs a feature-level topology-aware sorting strategy and a dual-channel graph neural network framework, enabling unified modeling of both feature and structure. Extensive experiments demonstrate the strong generalizability of TFSGFM. The source code is available at https://github.com/hedongxiao-tju/TFSGFM.

Fake news detection is a critical challenge in the digital age, as the fabricated content rapidly undermines public trust across diverse domains. Existing methods rely on domain-agnostic feature extractors, which struggle to decouple heterogeneous features and suffer from semantic entanglement, reducing cross-domain generalization. Moreover, static architectures with fixed knowledge representations cannot adapt to evolving misinformation patterns during events such as public health crises and political upheavals. Conventional multi-domain frameworks also incur high computational costs because of redundant parameter scaling, hindering near real-time deployment. To address these issues, herein, we propose a D ual-layer H eterogeneous E xpert network and dynamic M emory augmentation for multi-domain F ake N ews D etection (DHEM-FND). First, a heterogeneous expert network disentangles domain-specific semantics through three pathways: (i) a multiscale convolutional neural network capturing local–global patterns, (ii) emotion-aware multilayer perceptron detecting sentiment–context discrepancies, (iii) and self-attention module with statistical priors identifying stylistic fingerprints. Orthogonal constraints minimize feature leakage across domains. Second, a dynamic memory augmentation mechanism tracks evolving fake news tactics by maintaining hierarchical domain prototypes, incrementally updated using attention-based alignment. Third, sparse gating and parameter reuse strategies optimize efficiency by activating only the top-k relevant experts during inference while freezing 75% of backbone parameters. We validated the DHEM model using four real-world datasets: Twitter15, Twitter16, Weibo21, and FineFake. We achieved an average detection accuracy of 92.57%, while reducing cross-domain interference by 41.2%, improving detection speed by 17.2%. Thus, the superior performance of the proposed method in dealing with evolving fake news and scalability is demonstrated.

Background . Software vulnerabilities are increasing in complexity and scale, posing great security risks to many software systems. Extracting information about software vulnerabilities is a critical area of research that aims to identify and create a structured representation of vulnerability-related information. These structured data help software systems better understand vulnerabilities and provide security professionals with timely information to mitigate the impact of rapidly growing vulnerabilities while guiding future research to develop more secure systems. However, this process relies on the effectiveness of information extraction to transform manual vulnerability analysis from security experts to digital solutions. Despite its importance, the unique nature of vulnerability information and the fast pace at which machine learning-based extraction methods and techniques have evolved make it challenging to assess the current successes, failures, challenges, and opportunities within this research area. This study presents a systematic literature review aimed at clarifying this complex landscape. Methods . In this study, we conduct a systematic literature review (SLR) to explore existing research focusing on extracting information about software security vulnerabilities. We search for 829 primary studies on security vulnerability information extraction from seven widely used online digital libraries, focusing on top peer-reviewed journals and conferences published between 2001 and 2024. After applying our inclusion and exclusion criteria and the snowballing technique, we narrowed our selection to 87 studies for in-depth analysis and addressed four main research questions. We collect qualitative and quantitative data from each study, identifying 34 components such as research problems, methods, contributions, evaluation metrics, results, types of extracted vulnerability information, challenges, and limitations. We use meta-analysis, statistical machine learning, and text-mining techniques to identify themes, patterns, and trends across the primary studies and visualize findings. Result : The study provides an overview of the security vulnerability data landscape, identifies key resources, and guides efforts to improve vulnerability information extraction and analysis. The study finds a diverse landscape of learning algorithms used in security vulnerability information extraction, with Bidirectional Encoder Representations from Transformers (BERT), Long Short-term Memory (LSTM), and Support Vector Machine (SVM) being the most dominant. The study identifies key challenges, including feature engineering complexity, lack of a gold-standard corpus, preprocessing errors, generating accurate training data, addressing imbalanced data, multimodality fusion, and graph sparsity in security knowledge graphs. Insights for Future Research Directions . The study underscores the need for advanced extraction approaches, robust datasets, automated annotation methods, and advanced machine learning algorithms to improve the extraction of security vulnerability information. This study also suggests using large language models (LLMs) and transformer models to facilitate the automatic extraction of security-related words, terms, concepts, and phrases and introduce new filtering parameters for user requirements. We provide all our implementations; it can be found at https://bitbucket.org/slr-svie/vulnerability-information-extraction/src/master/ .

Interactive Recommendation (IR) has gained significant attention recently for its capability to quickly capture dynamic interest and optimize both short and long term objectives. IR agents are typically implemented through Deep Reinforcement Learning (DRL), because DRL is inherently compatible with the dynamic nature of IR. However, DRL is currently not perfect for IR. Due to the large action space and sample inefficiency problem, training DRL recommender agents is challenging. The key point is that useful features cannot be extracted as high-quality representations for the recommender agent to optimize its policy. To tackle this problem, we propose Contrastive Representation for Interactive Recommendation (CRIR). CRIR efficiently extracts latent, high-level preference ranking features from explicit interaction, and leverages the features to enhance users’ representation. Specifically, the CRIR provides representation through one representation network, and refines it through our proposed Preference Ranking Contrastive Learning (PRCL). The key insight of PRCL is that it can perform contrastive learning without relying on computations involving high-level representations or large potential action sets. Furthermore, we also propose a data exploiting mechanism and an agent training mechanism to better adapt CRIR to the DRL backbone. Extensive experiments have been carried out to show our method's superior improvement on the sample efficiency while training an DRL-based IR agent.

Graph Neural Networks (GNNs) have recently achieved significant success in several graph-related tasks. However, traditional GNNs and their variants are constantly limited by the implicit homophily, assuming neighboring nodes belong to the same class. This results in weak performance on heterophilic graphs where most nodes are linked to neighbors of different classes. Despite the numerous attempts to adequately deal with heterophily, most methods still use the uniform propagation aggregation mechanism. In this paper, we argue that identifying neighbors with different class labels and exploiting them individually is crucial for heterophilic GNNs. We then propose a simple and efficient novel co-training approach, EG-GCN, which uses group aggregation to handle homophilic and heterophilic neighbors separately. In EG-GCN, we first use an edge discriminator to classify edges and split the neighborhood of every node into two parts. We then apply group graph convolution to the divided neighborhoods to obtain node representations. During training, we continuously optimize the edge discriminator to improve neighborhood partition and use the node classification results to identify highly confident unlabeled nodes to expand the edge training set. This co-training strategy enables both components to enhance each other mutually. Extensive experiments demonstrate that EG-GCN significantly outperforms the state-of-the-art approaches.

KG-based fact verification verifies the truthfulness of claims by retrieving evidence graphs from the knowledge graph. The faithful inference chains, which are precise relation paths between the mentioned entities and evidence entities, retrieve precise evidence graphs addressing poor performance and weak logic for fact verification. Due to the diversity of relation paths, existing methods rarely extract faithful inference chains. To alleviate these issues, we propose Multi-view Heterogeneous Graph with Causal Intervention (MHGCI): (i) We construct a Multi-view Heterogeneous Graph enhancing relation path extraction from the view of different mentioned entities. (ii) We propose a self-optimizing causal intervention model to generate assistant entities mitigating the out-of-distribution problem caused by counterfactual relations. (iii) We propose a grounding method to extract evidence graphs from the KG by faithful inference chains. Experiments on the public KG-based fact verification dataset FactKG demonstrate that our model provides precise evidence graphs and achieves state-of-the-art performance.

For black-box attacks, most existing attack methods exhibit weak transferability due to the significant discrepancy between substitute model and victim model. We argue that the model-specific discriminative regions are a key factor causing overfitting to the source model. However, existing model augmentation methods focus on augmentations within a single domain, thereby restricting the diversity of the simulated models. In this paper, we present a novel model augmentation method named CSFD, which combines our proposed channel-wise scaling(CS) and frequency-random dropping(FD) to enhance the diversity of simulated model. Specifically, we first scale the image by channel with CS, which augments them in spatial domain. Then we randomly remove specific frequency patterns of the image with our FD, further augmenting the image in frequency domain by introducing a loss-preserving transformation. This enables us to fully exploit the properties of image in different domains and largely increases the diversity of simulated models. Additionally, we inject random noise perturbations into the sample, effectively exploring the decision boundary within an extended data distribution space. Extensive experiments on ImageNet dataset show that the proposed method has better performance than the existing methods.

In the realm of service recommendation, personalized systems are essential for addressing information overload and satisfying diverse user preferences. However, traditional models, bound by a static training-test framework, often fail to keep pace with the dynamic nature of evolving user interests and expanding service catalogs. To tackle these challenges, we introduce an innovative incremental recommendation strategy that employs graph neural network(GNN) fine-tuning and interest alignment. This method dynamically updates users' latest interests while aligning them with pertinent long-term interests, effectively preventing catastrophic forgetting and minimizing dependency on historical data. Our approach not only ensures responsiveness to the most recent user interactions but also preserves valuable prior interests, demonstrating exceptional adaptability and performance in dynamic service recommendation environments. Furthermore, our strategy is designed to be seamlessly integrated into various existing GNN-based recommendation models. Extensive experiments conducted on three industrial datasets demonstrate the effectiveness and robustness of our method, highlighting its practical applicability and superior performance in dynamic recommendation scenarios.

As privacy protection gains increasing importance, more models are being trained on edge devices and subsequently merged into the central server through Federated Learning (FL). However, current research overlooks the impact of network topology, physical distance, and data heterogeneity on edge devices, leading to issues such as increased latency and degraded model performance. To address these issues, we propose a new federated learning scheme on edge devices that called Federated Learning with Encrypted Data Sharing(FedEDS). FedEDS uses the client model and the model's stochastic layer to train the data encryptor. The data encryptor generates encrypted data and shares it with other clients. The client uses the corresponding client's stochastic layer and encrypted data to train and adjust the local model. FedEDS uses the client's local private data and encrypted shared data from other clients to train the model. This approach accelerates the convergence speed of federated learning training and mitigates the negative impact of data heterogeneity, making it suitable for application services deployed on edge devices requiring rapid convergence. Experiments results show the efficacy of FedEDS in promoting model performance.

Heterophily emerges as a critical challenge in Graph Anomaly Detection (GAD). Recent studies reveal that neighborhood distributions, rather than heterophily itself, are the fundamental factor for the expressive power of Graph Neural Networks (GNNs). However, two key challenges remain unresolved. First, the overlap in neighborhood distributions between anomalous and normal nodes poses significant difficulties in distinguishing them effectively. Second, the dispersion in neighborhood distributions within the same class prevents the application of a fixed aggregation strategy to accommodate the diverse patterns within the class. To tackle the aforementioned challenges, we propose a novel Graph Neural Network model called Neighborhood Adaptive Aggregation and Spectral Tuning (NAAST-GNN). Specifically, we first design a neighborhood adaptive aggregation module that adjusts the message passing mechanism based on the predicted probabilities for different node classes, ensuring that nodes from distinct classes but with similar neighborhood distributions derive unique aggregated neighborhood information. We then present a spectral tuning module that dynamically selects and combines spectral filters based on the predicted neighborhood distribution, ensuring adaptability to the diverse neighborhood distributions of nodes within the same class. Comprehensive experimental results demonstrate that our method outperforms state-of-the-art baselines.

The recent progress in data and modeling techniques has significantly improved Named Entity Recognition (NER) for well-structured texts and high-resource languages. However, Roman Urdu, written in Latin script, poses unique challenges due to its informal structure, code-switching tendencies, and lack of standardized resources. The existing NER dataset for Roman Urdu suffers from inconsistent annotations and undefined entity boundaries, while traditional deep-learning models struggle with informal and multilingual text. To address these challenges, this study makes several key contributions. First, we re-annotate the Roman Urdu dataset to ensure tagging consistency, adopting the BIOES tagging scheme for entity boundary representation. Second, we fine-tuned multilingual pre-trained language models (PLMs), including mBERT, MuRIL, and XLM-R. Finally, we introduce XLM-R-IDCNN-BiLSTM-H, a novel architecture that combines transformer-based embeddings with advanced neural components, including Iterated Dilated Convolutional Neural Networks (IDCNN) for local dependencies, Bidirectional Long Short-Term Memory (BiLSTM) networks for bidirectional context, and highway networks for smooth information flow. We employ a conditional random field (CRF) to obtain the optimal tag sequence for classification. According to experimental findings, the proposed approach delivers state-of-the-art performance, which yielded an F1 score of 84.26% on the re-annotated Roman Urdu dataset.

Graph Prompt Learning (GPL) has emerged as a promising paradigm that bridges graph pretraining models and downstream scenarios, mitigating label dependency and the misalignment between upstream pretraining and downstream tasks. Although existing GPL studies explore various prompt strategies, their effectiveness and underlying principles remain unclear. We identify two critical limitations: (1) Lack of consensus on underlying mechanisms: Despite current GPLs have advanced the field, there is no consensus on how prompts interact with pretrained models, as different strategies intervene at varying spaces within the model, i.e., input-level, layer-wise, and representation-level prompts. (2) Limited scenario adaptability: Most methods fail to generalize across diverse downstream scenarios, especially under data distribution shifts (e.g., homophilic-to-heterophilic graphs). To address these issues, we theoretically analyze existing GPL approaches and reveal that representation-level prompts essentially function as fine-tuning a simple downstream classifier, proposing that graph prompt learning should focus on unleashing the capability of pretrained models, and the classifier should adapt to downstream scenarios. Based on our findings, we propose UniPrompt, a novel GPL method that adapts any pretrained models, unleashing the capability of pretrained models while preserving the input graph. Extensive experiments demonstrate that our method can effectively integrate with various pretrained models and achieve strong performance across in-domain and cross-domain scenarios.

Logical rules have been widely used for expressing schema knowledge in various practical applications. It is infeasible to handcraft rules from large knowledge graphs (KGs) and thus many methods have been proposed for learning rules automatically from KGs. However, it is largely ignored how to extract rules in a (target) KG from rules that already exist in some other (source) KGs. In this paper, we propose a framework for KG rule learning based on transfer learning. A major challenge for establishing such a framework is that a suitable alignment mechanism is required for mapping certain subgraph structures between predicates in the source KG and the target KG. Hence, our framework provides a new method for predicate mapping based on graph-structural similarity. The proposed framework can be used as a standalone rule learner but more importantly, it paves a new way for enhancing the state-of-the-art rule learners for large KGs. Extensive experiments are conducted to evaluate the new approach to rule learning, which shows that rules in smaller KGs can be effectively transferred to a large KG.

Abstract As the core operator of regular path query (RPQ), the Kleene closure is essentially a recursive operation based on predicates. It has more expressive power than first-order logic, but its algorithmic complexity is an NP-hard problem and expensive to execute. Most approaches implement Kleene closure via online recursive queries (inefficient for large-scale data), yet significant reusable intermediate results emerge during execution. This paper proposes an efficient query optimization method based on offline preprocessing and tree-structured strategies to address the challenges of NP-hard complexity and low execution efficiency caused by the recursive nature of Kleene closure RPQs (KRPQs). The core contributions of this work include: (i) recursive index tree: by precomputing and materializing intermediate results of predicate paths offline, Kleene closure queries are transformed into non-recursive SPARQL queries. This allows direct retrieval of answer branches from the preconstructed index tree, eliminating redundant computations and reducing online complexity. (ii) Selectivity-driven dynamic optimization: a heuristic cost model estimates intermediate result sizes to guide query plan generation, minimizing redundant operations. (3) Query decomposition tree for nested KRPQs: a hierarchical decomposition algorithm processes nested Kleene closures layer-by-layer from inner to outer expressions, overcoming limitations of existing methods in handling complex nested structures. Experiments on large-scale RDF datasets demonstrate that the method reduces average response times by 50%–70% for single-predicate and expression-based Kleene closure queries compared to systems like Virtuoso, Jena, and KRPQ, with over 60% improvement for complex nested queries. Additionally, multithreaded preprocessing and path pool optimizations achieve up to a 76$\times $ reduction in index construction time. Experiments show that the method can significantly reduce the time of KRPQs based on large RDF graphs.

ABSTRACT Sarcasm detection is a critical and challenging task in sentiment analysis, particularly for low‐resource languages like Urdu, where limited annotated data, linguistic complexity, and subtle contextual cues hinder accurate classification. Traditional machine learning methods often fail to capture the nuanced and often contradictory nature of sarcastic expression. To address these challenges, this paper presents a comprehensive and computationally efficient framework for Urdu sarcasm detection. We first mitigate severe class imbalance through strategic down‐sampling and back‐translation‐based data augmentation. We then conduct extensive benchmarking of traditional deep learning architectures against fine‐tuned pre‐trained language models, including multilingual, monolingual, and Twitter‐specific variants. Building on these insights, we propose DeepSarc, a novel hybrid model that integrates the contextual embeddings from XLM‐T, the multi‐scale feature extraction capabilities of dilated convolutional neural networks (DCNNs), and the sequential dependency modeling of bidirectional long short‐term memory (BiLSTM) networks. While slightly more computationally intensive than simpler alternatives, DeepSarc achieves a state‐of‐the‐art F1‐score, significantly outperforming existing approaches. Our results establish a new benchmark for sarcasm detection in low‐resource languages and provide a scalable, high‐performance framework adaptable to diverse linguistic contexts.

The non-cooperative game between rational vehicle users will generate unnecessary costs, manifested as the gap between user equilibrium (UE) and system optimal (SO). This issue primarily arises due to the competition or the untrusted collaboration among vehicles. The traditional marginal cost pricing (MCP) method is constrained by factors such as vehicle density and communication protocols, resulting in suboptimal performance. In this paper, the immutability of Hashgraph is leveraged to enable trusted services in the Internet of Vehicles (IoV), transforming non-cooperative games into a global optimization problem, while a two-phase optimization method is proposed to achieve low-cost services. Firstly, this paper simplifies the process of determining consensus timestamps for Hashgraph and constrains the actions of participants through immutability, thereby guaranteeing trusted services more efficiently. Subsequently, this paper systematically analyzes the key factors affecting travel and network service costs to optimize them in turn. Specifically, regarding the travel costs, this paper introduces a segment shielding method in trusted scenarios to avoid Braess’s paradox. As for the network service costs of data sharing, this paper presents a latency-sensitive dynamic programming method to integrate each server’s status to optimize resource scheduling. When the vehicle density is 400, the proposed method reduces the total cost by 20.920% and improves the QoE by 122.535%. The advantages become more significant as vehicle density increases.

Traditional DRAM-based memory solutions face challenges, including high energy consumption and limited scalability. Non-Volatile Memory (NVM) offers low energy consumption and high scalability. However, security challenges, particularly data remanence vulnerabilities, persist. Prevalent methods such as the Bonsai Merkle Tree (BMT) are employed to ensure data security. However, the consistency requirements for integrity tree updates have led to performance issues. It is observed that compared to a secure NVM system without persistent secure metadata, the average overhead for updating and persisting the BMT root with persistent secure metadata is as high as 2.48 times. Therefore, this paper aims to mitigate these inefficiencies by leveraging the principle of memory access locality. We propose the Dynamic Construction and Adjustment of Skewed Bonsai Merkle Tree (DCAS-BMT). The DCAS-BMT is dynamically built and continuously adjusted at runtime according to access weights, ensuring frequently accessed memory blocks reside on shorter paths to the root node. This reduces the verification steps for frequently accessed memory blocks, thereby lowering the overall cost of memory authentication and updates. Experimental results using the USIMM memory simulator demonstrate that compared to the widely used BMT approach, the DCAS-BMT scheme shows a performance improvement of 34.1%.

Contrastive Learning (CL) has emerged as a popular self-supervised representation learning paradigm that has been shown in many applications to perform similarly to traditional supervised learning methods. A key component of CL is mining the latent discriminative relationships between positive and negative samples and using them as self-supervised labels. We argue that this discriminative contrastive task is, in essence, similar to a classification task, and the “either positive or negative” hard label sampling strategies are arbitrary. To solve this problem, we explore ideas from data distillation, which considers probabilistic logit vectors as soft labels to transfer model knowledge. We attempt to abandon the classical hard sampling labels in CL and instead explore self-supervised soft labels. We adopt soft sampling labels that are extracted, without supervision, from the inherent relationships in data pairs to retain more information. We propose a new self-supervised graph learning method, Distill and Contrast (D&C), for learning representations that closely approximate natural data relationships. D&C extracts node similarities from the features and structures to derive soft sampling labels, which also eliminate noise in the data to increase robustness. Extensive experimental results on real-world datasets demonstrate the effectiveness of the proposed method.

Traditional recommendation systems focus on maximizing user satisfaction by suggesting their favorite items. This user-centric approach may lead to unfair exposure distribution among the providers. On the contrary, a provider-centric design might become unfair to the users. Therefore, this paper proposes a re-ranking model FairSort1to find a trade-off solution among user-side fairness, provider-side fairness, and personalized recommendations utility. Previous works habitually treat this issue as a knapsack problem, incorporating both-side fairness as constraints. In this paper, we adopt a novel perspective, treating each recommendation list as a runway rather than a knapsack. In this perspective, each item on the runway gains a velocity and runs within a specific time, achieving re-ranking for both-side fairness. Meanwhile, we ensure the Minimum Utility Guarantee for personalized recommendations by designing a Binary Search approach. This can provide more reliable recommendations compared to the conventional greedy strategy based on the knapsack problem. We further broaden the applicability of FairSort, designing two versions for online and offline recommendation scenarios. Theoretical analysis and extensive experiments on real-world datasets indicate that FairSort can ensure more reliable personalized recommendations while considering fairness for both the provider and user.

With the rapid development of the Internet of Vehicles (IoV), vehicles will generate massive data and computation demands, necessitating computation offloading at the edge. However, existing research faces challenges in efficiency and trust. In this paper, we explore the IoV computation offloading from both user and edge facility provider perspectives, working to optimize the quality of experience (QoE), load balancing, and success rate based on challenges to efficiency and trust. First, two vehicle interconnection models are constructed to extend the linkable range of intra-road and inter-road vehicles while considering the maximum link time constraint. Then, a dynamic planning method is proposed, combining the reputation and feedback mechanisms, which can schedule edge resources online based on the cumulative computation latency of each service side, reliability value, and historical behavior. These two phases further improve the efficiency of edge services. Subsequently, blockchain is combined to optimize the trust problem of edge collaboration, and an edge-limited Byzantine fault tolerance local consensus mechanism is proposed to optimize consensus efficiency and ensure the reliability of edge services. Finally, this paper conducts dynamic experiments on real-world datasets, verifying the effectiveness of the proposed algorithm and models in multiple vehicle density datasets and experimental scenarios.

In software development, the raw requirements proposed by users are frequently incomplete, which impedes the complete implementation of software functionalities. With the emergence of large language models, the exploration of generating software through user requirements has attracted attention. Recent methods with the top-down waterfall model employ a questioning approach for requirement completion, attempting to explore further user requirements. However, users, constrained by their domain knowledge, result in a lack of effective acceptance criteria during the requirement completion, failing to fully capture the implicit needs of the user. Moreover, the cumulative errors of the waterfall model can lead to discrepancies between the generated code and user requirements. The Agile methodologies reduce cumulative errors of the waterfall model through lightweight iteration and collaboration with users, but the challenge lies in ensuring semantic consistency between user requirements and the code generated by the agent. To address these challenges, we propose AgileGen, an agile-based generative software development through human-AI teamwork. Unlike existing questioning agents, AgileGen adopts a novel collaborative approach that breaks free from the constraints of domain knowledge by initiating the end-user perspective to complete the acceptance criteria. By introducing the Gherkin language, AgileGen attempts for the first time to use testable requirement descriptions as a bridge for semantic consistency between requirements and code, aiming to ensure that software products meet actual user requirements by defining user scenarios that include acceptance criteria. Additionally, we innovate in the human-AI teamwork model, allowing users to participate in decision-making processes they do well and significantly enhancing the completeness of software functionality. To ensure semantic consistency between requirements and generated code, we derive consistency factors from Gherkin to drive the subsequent software code generation. Finally, to improve the reliability of user scenarios, we also introduce a memory pool mechanism, collecting user decision-making scenarios and recommending them to new users with similar requirements. AgileGen, as a user-friendly interactive system, significantly outperformed existing best methods by 16.4% and garnered higher user satisfaction.

Textual vulnerability descriptions (TVDs) document phrase-based concepts of security vulnerabilities, such asVulnerability TypeandAttack Vector. For example, a TVD might state, “Abuffer overflow vulnerability(Vulnerability Type) in the image processing library allowsremote code execution(Impact).” The manual creation of TVDs often leads to missing phrase-based concepts, hindering severity and exploitability assessments. Large language models (LLMs) are used to complete these missing phrase-based concepts, but they suffer from hallucinations, generating plausible yet inaccurate completion that burden developers. Existing methods mitigate hallucinations by aligning results with knowledge bases, which fails when hallucinated predictions appear believable due to knowledge base similarity. To address this, we propose leveraging 1-hop relationships between phrase-based concepts in a TVD to filter hallucinated predictions. Our approach consists of three modules: 1-hop relationship generation, which enhances prediction diversity by introducing TVD perturbations to counter the LLM’s tendency to generate outputs similar to the context; semantic augmentation using CWE descriptions to enrich the 1-hop relationships and phrase-based concepts, addressing the challenge of capturing concise technical semantics; and chain-of-thought (CoT) reasoning for filtering, which guides the LLM to articulate and verify logical connections, ensuring contextually accurate and consistent 1-hop relationships. Experiments on a TVD dataset show that our method reduces hallucinations, improves completion accuracy from 49% to 62%, and enhances the reliability of LLM-generated phrase-based concepts.

The emergence of the mobile edge computing paradigm enables the deployment of microservices on edge servers, which greatly improves the quality of services and reduces network transmission costs. However, due to limited computing and storage resources, an individual edge server can host only a limited number of microservice instances. Moreover, user mobility often results in uneven distribution of service requests in mobile edge computing systems. To this end, it is a key problem to dispatch microservice requests to appropriate edge servers to minimize the average service response time. Current solutions to this problem rely on centralized methods and suffer from serious problems of single point of failure, error-proneness, difficult expansion, low robustness, etc. To resolve these problems, this paper proposes a decentralized game-theoretic approach for dispatching microservice requests effectively and efficiently in mobile edge computing systems. Specifically, we formulate the request dispatching problem as a decentralized non-cooperative game and propose a decentralized request dispatching algorithm that can find the Nash equilibrium through finite iterations. We conduct a series of experiments to demonstrate that our approach beats benchmarking approaches with close-to-optimal performance and high efficiency measured by convergence time.

The increasing prevalence of smart services in daily life drives the rapid emergence of service ecosystems across various domains such as E-commerce, cloud manufacturing, and crowdsourcing. Evaluating the utility of these ecosystems is challenging due to complex characteristics like diverse crowd intelligence, cascading service network effects, and the interplay of individual interests. To address these challenges, this study introduces an innovative utility evaluation model that integrates individual and systemic factors with multidimensional metrics, reconciling micro and macro perspectives. This model effectively addresses potential conflicts between individual and systemic benefits, supporting the continuous learning and evolution of agents within service ecosystems. It employs value entropy to precisely model and interpret complex nonlinear emergence phenomena. The model's universal framework offers high customizability and broad applicability, facilitating specific adaptations across different service ecosystem scenarios. Additionally, a visual multi-agent system simulation tool has been developed to adjust agent attributes and cooperative topologies, allowing for the observation of system responses. Empirical validation confirms the model's accuracy and provides a mechanistic explanation for emergence phenomena in human societies. The findings demonstrate the model's high coordination and effectiveness in managing both linear and nonlinear characteristics, presenting a powerful and flexible tool for utility evaluation in service ecosystems. Our code is available at https://github.com/yxn9191/value_entropy.

Augmenting missing key aspects in Textual Vulnerability Descriptions (TVDs) is crucial for effective vulnerability analysis. For instance, in TVDs, key aspects includeAttack Vector,Vulnerability Type, among others. These key aspects help security engineers understand and address the vulnerability in a timely manner. For software with a large user base (non-long-tail software), augmenting these missing key aspects has significantly advanced vulnerability analysis and software security research. However, software instances with a limited user base (long-tail software) often get overlooked due to inconsistency software names, TVD limited avaliability, and domain-specific jargon, which complicates vulnerability analysis and software repairs. In this paper, we introduce a novel software feature inference framework designed to augment the missing key aspects of TVDs for long-tail software. Firstly, we tackle the issue of non-standard software names found in community-maintained vulnerability databases by cross-referencing government databases with Common Vulnerabilities and Exposures (CVEs). Next, we employ Large Language Models (LLMs) to generate the missing key aspects. However, the limited availability of historical TVDs restricts the variety of examples. To overcome this limitation, we utilize the Common Weakness Enumeration (CWE) to classify all TVDs and select cluster centers as representative examples. To ensure accuracy, we present Natural Language Inference (NLI) models specifically designed for long-tail software. These models identify and eliminate incorrect responses. Additionally, we use a wiki repository to provide explanations for proprietary terms. Our evaluations demonstrate that our approach significantly improves the accuracy of augmenting missing key aspects of TVDs for log-tail software from 0.27 to 0.56 (+107%). Interestingly, the accuracy of non-long-tail software also increases from 64% to 71%. As a result, our approach can be useful in various downstream tasks that require complete TVD information.

Graph Neural Networks (GNNs) have received widespread attention and applications due to their excellent performance in graph representation learning. Most existing GNNs can only aggregate 1-hop neighbors in a GNN layer, so they usually stack multiple GNN layers to obtain more information from larger neighborhoods. However, many studies have shown that model performance experiences a significant degradation with the increase of GNN layers. In this paper, we first introduce the concept of distinguishability of class to indirectly evaluate the learned node representations, and verify the positive correlation between distinguishability of class and model performance. Then, we propose a Graph Neural Network guided by Distinguishability of class (Disc-GNN) to monitor the representation learning, so as to learn better node representations and improve model performance. Specifically, we first perform inter-layer filtering and initial compensation based on Local Distinguishability of Class (LDC) in each layer, so that the learned node representations have the ability to distinguish different classes. Furthermore, we add a regularization term based on Global Distinguishability of Class (GDC) to achieve global optimization of model performance. Extensive experiments on six real-world datasets have shown that the competitive performance of Disc-GNN to the state-of-the-art methods on node classification and node clustering tasks.

Graph contrastive learning (GCL) has attracted considerable attention because it can self-supervisedly extract low-dimensional representation of graph data. InfoNCE-based loss function is widely used in graph contrastive learning, which pulls the representations of positive pairs close to each other and pulls the representations of negative pairs away from each other. Recent works mainly focus on designing new augmentation methods or sampling strategies. However, we argue that the widely used InfoNCE-based methods may contain an implicit conflict which seriously confuses models when learning from negative pairs. This conflict is engendered by the encoder's message-passing mechanism and the InfoNCE loss function. As a result, the learned representations between negative samples cannot be far away from each other, compromising the model performance. To our best knowledge, this is the first time to report and analysis this conflict of GCL. To address this problem, we propose a simple but effective method called Partial ignored Graph Contrastive Learning (PiGCL). Specifically, PiGCL first dynamically captures the conflicts during training by detecting the gradient of representation similarities. It then enables the loss function to ignore the conflict, allowing the encoder to adaptively learn the ignored information without self-supervised samples. Extensive experiments demonstrate the effectiveness of our method.

Detecting OOD inputs is crucial to deploy machine learning models to the real world safely. However, existing OOD detection methods require an in-distribution (ID) dataset to retrain the models. In this paper, we propose a Deep Generative Models (DGMs) based transferable OOD detection that does not require retraining on the new ID dataset. We first establish and substantiate two hypotheses on DGMs: DGMs exhibit a predisposition towards acquiring low-level features, in preference to semantic information; the lower bound of DGM's log-likelihoods is tied to the conditional entropy between the model input and target output. Drawing on the aforementioned hypotheses, we present an innovative image-erasing strategy, which is designed to create distinct conditional entropy distributions for each individual ID dataset. By training a DGM on a complex dataset with the proposed image-erasing strategy, the DGM could capture the discrepancy of conditional entropy distribution for varying ID datasets, without re-training. We validate the proposed method on the five datasets and show that, without retraining, our method achieves comparable performance to the state-of-the-art group-based OOD detection methods. The project codes will be open-sourced on our project website.

In graph anomaly detection (GAD), the fact that anomalous nodes usually exhibit high heterophily, while most Graph Neural Networks (GNNs) have homophily assumptions, leads to poor performance. Many studies have attempted to solve this problem by employing a set of graph filters covering various frequencies. Their ultimate goal is to design the most appropriate spectral filter to capture the complex signals generated by normals and anomalies. The critical aspect lies in the fusion of information from filters with different frequency response functions. However, existing methods lack a clear indicator to guide the fusion of information at different frequencies. In this paper, we find that local homophily is a valuable metric for assessing the weights of high- and low-frequency information at the node level, and explicitly point out that the accuracy of local homophily is positively correlated with the accuracy of anomaly detection. Moreover, we unveil the phenomenon of camouflage in anomalous, wherein these nodes disguise themselves by making their features resemble those of surrounding normals.

Time-Sensitive Question Answering (TSQA) is to answer questions qualified for a certain timestamp based on the given document. It is split into easy and hard modes depending on whether the document contain time qualifiers mentioned in the question. While existing models have performed well on easy mode, their performance is significant reduced for answering hard time-sensitive questions, whose time qualifiers are implicit in the document. An intuitive idea is to match temporal events in the given document by treating time-sensitive question as a temporal event of missing objects. However, not all temporal events extracted from the document have explicit time qualifiers. In this paper, we propose an Event-AL framework, in which a graph pruning model is designed to locate the timespan of implicit temporal events by capturing temporal relation between events. Moreover, we present an abductive reasoning module to determine proper objects while providing explanations. Besides, as the same relation may be scattered throughout the document in diverse expressions, a relation-based prompt is introduced to instructs LLMs in extracting candidate temporal events. We conduct extensive experiment and results show that Event-AL outperforms strong baselines for hard time-sensitive questions, with a 12.7% improvement in EM scores. In addition, it also exhibits great superiority for multi-answer and beyond hard time-sensitive questions.

In the real world, it can be challenging to annotate a large-scale dataset for all medical images, making few-shot medical image classification an important task. The latest advancements in pre-trained vision-language models as CLIP have demonstrated excellent performance in zero-shot natural image recognition and show advantages in medical applications. However, we have found that deploying such models in practical applications faces challenges in terms of engineering effort. It requires specialized medical domain knowledge and is time-consuming, as even slight variations in wording can have a significant impact on performance. Inspired by recent research on prompt learning in the field of Natural Language Processing (NLP), we propose a simple approach called Prompt Learner (PoLe) for automating the design of prompts in pre-trained vision-language models. This is a simple method specifically designed to fine-tune vision-language models, similar to CLIP, for downstream image recognition. In particular, PoLe models context tokens using continuous vectors that can automatically learn from medical images, thus avoiding the tedious process of handcrafting prompt engineering. Additionally, this approach maintains the frozen state of the large-scale pre-trained parameters, saving computational resources. Through extensive experiments on 5 medical image datasets, we have demonstrated that PoLe surpasses manually designed prompts with just one or two shots, and further training with more shots significantly improves the performance of image classification. For instance, when trained with 16 shots, the average improvement is approximately 20% (with a maximum improvement of over 31%). PoLe effectively transforms CLIP into a powerful few-shot learner. In terms of recognition performance, adjusting the CLIP model using PoLe yields better results than manually designed prompts for CLIP. When enhancing CLIP, PoLe demonstrates stronger learning capabilities compared to other few-shot learners such as linear probes. Furthermore, it outperforms CLIP models assisted by ChatGPT on most datasets. This indicates that PoLe possesses significant adaptability in the field of medical image analysis.

Financial institutions may be subject to financial fraud by malicious users because of the large amount of transaction data and sensitive user information involved. Therefore, it is crucial to design a machine learning model that can detect abnormal data in financial institutions. However, with the development of the economy and technology, the massive amount of user-generated data is distributed among various financial institutions, and how to enable multiple financial institutions to collaborate on anomalous data detection has become a new challenge. In this paper, we propose a blockchain-based federated learning architecture to assist multiple financial institutions to collaborate on anomaly detection. First, anomaly detection models are trained locally without sharing local data, which effectively protects data privacy. Second, the architecture introduces a differential privacy algorithm to protect data security in communication. Finally, to avoid communication bottlenecks that threaten data security, the architecture employs the aperiodic aggregation algorithm in which clients collaborate to reduce communication costs. Experimentally, a large number of experiments are conducted using three datasets to evaluate the proposed architecture. The experimental results show that the architecture is effective in detecting anomalous data and reducing communication costs.

Multiple-choice question answering (MCQA) becomes particularly challenging when all choices are relevant to the question and are semantically similar. Yet this setting of MCQA can potentially provide valuable clues for choosing the right answer. Existing models often rank each choice separately, overlooking the context provided by other choices. Specifically, they fail to leverage the semantic commonalities and nuances among the choices for reasoning. In this paper, we propose a novel MCQA model by differentiating choices through identifying and eliminating their commonality, called DCQA. Our model captures token-level attention of each choice to the question, and separates tokens of the question attended to by all the choices (i.e., commonalities) from those by individual choices (i.e., nuances). Using the nuances as refined contexts for the choices, our model can effectively differentiate choices with subtle differences and provide justifications for choosing the correct answer. We conduct comprehensive experiments across five commonly used MCQA benchmarks, demonstrating that DCQA consistently outperforms baseline models. Furthermore, our case study illustrates the effectiveness of the approach in directing the attention of the model to more differentiating features.

Syntactically controlled paraphrase generation is to generate diverse sentences that have the same semantics as the given original sentence but conform to the target syntactic structure. An optimal opportunity to enhance diversity is to make word substitutions during rephrasing based on syntactic control. Existing unsupervised methods have made great progress in syntactic control, but the generated paraphrases rarely have substitutions due to the limitation of training data. In this paper, we propose a Diversity syntactically controlled Paraphrase generation framework (DiPara), in which a novel training strategy is designed to obtain semantic sentences while using the given sentence as training objects. As diverse words vary the syntactic structure around them, we propose a phrase-aware attention mechanism to capture the syntactic structure associated with the current word. To achieve it, the linearized triple sequence is introduced to represent structure singly. Experiment results on two datasets show that DiPara outperforms strong baselines, especially diversity (Self-BLEU4) is improved by 10.18% in ParaNMT-Small.

Social intelligence is essential for understanding complex human expressions and social interactions.While large multimodal models (LMMs) have demonstrated remarkable performance in social intelligence question answering (SIQA), they are still inclined to generate responses relying on language priors and ignoring the relevant context due to the dominant prevalence of text-based data in the pretraining stage.To interpret the aforementioned language bias of LMMs, we employ a structure causal model and posit that counterfactual reasoning can mitigate the bias by avoiding spurious correlations between LMMs' internal commonsense knowledge and the given context.However, it is costly and challenging to construct multimodal counterfactual samples.To tackle the above challenges, we propose an output Distribution Calibration network with Virtual Counterfactual (DCVC) data augmentation framework.DCVC devises a novel output distribution calibration network to mitigate the impact of negative language biases while preserving beneficial priors.Perturbations are introduced to the output distributions of LMMs to simulate the distribution shifts from counterfactual manipulations of the context, which is employed to construct counterfactual augmented data virtually.Experiments on multiple datasets demonstrate the effectiveness and generalizability of our proposed method.

With the rapid development of prompt tuning technologies and the emergence of prompt-as-a-service platforms, the security of prompt service has attracted the attention of researchers. Deep neural networks face susceptibility to various adversaries. A particularly malicious altering model behavior is backdoor attack, where model predictions exhibit divergence in the presence of specific triggers in inputs. We concentrate on backdoor attacks in code intelligence models. In this paper, we aim to propose a novel backdoor attack scenario target the security of prompt service. In traditional backdoor scenario, the victim model is fine-tuned on the downstream poisoned dataset to establish a shortcut between the trigger and the target label. This scenario suffers from flaws such as rare words inserted into code snippets resulting in alterations to the code’ s semantics. And only those attackers who are familiar with specific triggers can launch an attack. Instead of introducing extra rare words to code snippets, our attack scenario choose to employ prompt itself as the triggers. We evaluate our methods on three popular code intelligence tasks, including code defect detection, clone detection and code summarization. Experimental results indicate that our methods can achieve an attack success rate of 95% to 99% with almost no sacrifice in accuracy on the original task, exhibiting an average performance degradation of less than 1%. We hope exposed potential security risks hidden in prompt service can raise awareness among researchers.

With the rapid development of mobile internet, mobile data traffic has increased substantially, and multimedia services have gradually occupied a large proportion of mobile network traffic. Using the Device-to-Device (D2D) technique for multimedia content sharing is considered as an effective solution to address large-scale user requests. D2D communication technology refers to direct data transmission between two adjacent users by reusing the spectrum resources of cellular users, effectively improving system throughput and spectrum resource utilization. Currently, most literature on D2D content sharing focuses on cache strategy research, while neglecting the design of reasonable content sharing mechanism. This paper focuses on the D2D service supply problem in content sharing scenarios and designs a stable sharing strategy by jointly considering power control and spectrum allocation. The power control and spectrum allocation problems for D2D users and cellular users are modeled as a minimum cost flow problem in graph theory and solved using the minimum cost flow algorithm. To ensure the stability of the content sharing system, a stable matching algorithm based on the delay acceptance algorithm is proposed for requesters and providers, considering preference factors. A series of experiments show that the proposed algorithm has good fairness and execution efficiency.

The governance of service ecosystem needs to balance efficiency and fairness to promote the sustainable development of the system, making it an important topic. However, service entities in the service ecosystem have autonomy and engage in dynamic game with governance strategies, leading to governance challenges. To address this issue, we model the governance process as a repeated sequential game between the governance algorithm and service entities, and propose a novel two-level learning algorithm. This algorithm considers the learning evolution of service entities and the co-evolutionary of the governance algorithm, using reinforcement learning to learn effective governance strategies while also considering the response function of service entities to balance efficiency and fairness. Combining the ideas of bilevel optimization and online learning, this algorithm effectively balances exploration (understanding the service entities’ responses function) and exploitation (choosing efficient actions). We apply the algorithm to a classic service ecosystem, the ride-hailing service system, and empirically demonstrate its effectiveness in the governance task of order dispatching.

With the rapid development of mobile communication and Internet of Things (IoT) technologies, smart mobile devices such as portable and sensor devices have been widely used in our daily lives. However, their compact size and limited energy capacity inherently hinder their ability to efficiently handle computation-intensive tasks within acceptable timeframes. To tackle this challenge, computation offloading has emerged as a pivotal solution. Computation offloading can significantly reduce the response time and energy consumption for mobile devices executing such tasks. However, it also bring some challenges, notably the risk of compromising user privacy. In this paper, we prioritize user privacy alongside considerations of service delay and energy consumption, and model the offloading decision problem as a multi-objective optimization problem. We employ an enhanced multi-objective bat algorithm to identify Pareto front solutions, balancing the diverse objectives effectively. Our experimental validation confirms the feasibility and efficacy of the proposed method, offering a promising avenue for addressing the complexities of computation offloading in mobile edge computing systems.

With the increasing prevalence of smart services in our daily lives, service ecosystems are rapidly emerging across various domains, including E-commerce, cloud manufacturing, and crowdsourcing, among others. However, evaluating the utility of service ecosystems presents several challenges due to their complex characteristics, including the diversity in crowd intelligence, the cascading effects of service networks, and the interest game among individuals. Current evaluation methods are limited in providing a comprehensive system-wide view, highlighting various shortcomings. To address these challenges, this article introduces a multidimensional and integrated model for evaluating service utility, comprising three key components: (1) The individual utility model is built to facilitate the transition from specific scenarios to general evaluations. (2) Value entropy is introduced to assess the impact of nonlinear emergence on system utility. (3) The system utility model is used to enable the integration of both linear and nonlinear factors within the system. Furthermore, a crowdsourcing service platform is taken as an example to verify the effectiveness of the proposed model. The results show that the proposed model is effective in providing new means and ideas for evaluating and analyzing service ecosystems.

Data service marketplaces are pivotal elements of an intelligent society as they bridge user requirements and the realization of value from data services. Ensuring the efficient circulation of data services and driving industrial chains to create value within data service marketplaces has become a core issue in developing intelligent societies, posing an urgent concern for researchers and governments. Within the context of service ecosystems, individual data services can no longer meet the requirements of users and organizations thus moving towards convergence, which brings new governance challenges both economically and technically. This paper comprehensively explores the governance of data service marketplaces by combining insights from service ecosystems and data service value. It begins by introducing data service marketplaces. Data servitization and standardized workflows are driving the massive expansion of data services marketplaces. Then it analyzes the complexity of marketplaces under service ecosystems. Furthermore, the governance issues and research themes of data service marketplaces were elaborated at macro, medium, and micro scales. This paper provides new ideas for the value creation and development of data service marketplaces.

The Conversational Recommendation System (CRS) aims to capture user dynamic preferences and provide item recommendations based on multi-turn conversations. However, effectively modeling these dynamic preferences faces challenges due to conversational limitations, which mainly manifests as limited turns in a conversation (quantity aspect) and low compliance with queries (quality aspect). Previous studies often address these challenges in isolation, overlooking their interconnected nature. The fundamental issue underlying both problems lies in the potential abrupt changes in user preferences, to which CRS may not respond promptly. We acknowledge that user preferences are influenced by temporal factors, serving as a bridge between conversation quantity and quality. Therefore, we propose a more comprehensive CRS framework called Time-aware User-preference Tracking for Conversational Recommendation System (TUT4CRS), leveraging time dynamics to tackle both issues simultaneously. Specifically, we construct a global time interaction graph to incorporate rich external information and establish a local time-aware weight graph based on this information to adeptly select queries and effectively model user dynamic preferences. Extensive experiments on two real-world datasets validate that TUT4CRS can significantly improve recommendation performance while reducing the number of conversation turns.

Graph Neural Networks (GNNs), known for their effective graph encoding, are extensively used across various fields. Graph self-supervised pre-training, which trains GNN encoders without manual labels to generate high-quality graph representations, has garnered widespread attention. However, due to the inherent complex characteristics in graphs, GNNs encoders pre-trained on one dataset struggle to directly adapt to others that have different node feature shapes. This typically necessitates either model rebuilding or data alignment. The former results in non-transferability as each dataset need to rebuild a new model, while the latter brings serious knowledge loss since it forces features into a uniform shape by preprocessing such as Principal Component Analysis (PCA). To address this challenge, we propose a new Feature-Universal Graph contrastive pre-training strategy (FUG) that naturally avoids the need for model rebuilding and data reshaping. Specifically, inspired by discussions in existing work on the relationship between contrastive Learning and PCA, we conducted a theoretical analysis and discovered that PCA's optimization objective is a special case of that in contrastive Learning. We designed an encoder with contrastive constraints to emulate PCA's generation of basis transformation matrix, which is utilized to losslessly adapt features in different datasets. Furthermore, we introduced a global uniformity constraint to replace negative sampling, reducing the time complexity from $O(n^2)$ to $O(n)$, and by explicitly defining positive samples, FUG avoids the substantial memory requirements of data augmentation. In cross domain experiments, FUG has a performance close to the re-trained new models. The source code is available at: https://github.com/hedongxiao-tju/FUG.

Our research reveals that multi-interest recommendation models in the matching stage tend to exhibit an under-clustered item embedding space, which leads to a low discernibility between items and hampers item retrieval. This highlights the necessity for item embedding enhancement. However, item attributes, which serve as effective side information for enhancement, are either unavailable or incomplete in many public datasets due to the labor-intensive nature of manual annotation tasks. This dilemma raises two meaningful questions: 1. Can we bypass manual annotation and directly simulate complete attribute information from the interaction data? And 2. If feasible, how can we simulate attributes with high accuracy and low complexity in the matching stage?

With the development of 5G technology, ultradense cellular networks are becoming a trend, while the deployment of large-area and high-density base stations (BSs) will bring new energy consumption problems. In this article, we explore the energy consumption of the edge demand response (EDR) from the two perspectives of edge users and edge facility providers. While guaranteeing the essential Quality of Experience (QoE), we try to improve the load balancing of edge servers and reduce system energy consumption. On the edge facility provider side, we mainly consider the physical machine turn-on problem and resource allocation of the two-phase EDR. On the user side, we start with user cost reduction and subchannel power allocation to emphasize the QoE. First, considering the density impacts of users and edge servers, we take inspiration from the classical PageRank algorithm and propose a method to calculate the edge nodes’ weights and, thus, determine the infrastructure’s state. Subsequently, combining distance factors, we design a subchannel power allocation method based on dynamic planning for 5G power-domain multiplexing nonorthogonal multiple access (PDM-NOMA). More importantly, based on the above work, we optimize the two-phase EDR process based on the upper confidence bound (UCB) algorithm of the multiarmed bandit (MAB) algorithm framework and dynamic planning. We compare the proposed QEL-UCB algorithm with two classical and five state-of-the-art algorithms on a real-world data set. The experimental results demonstrate that the proposed method improves by 18.52% in load balancing and reduces by 18.59% in energy consumption, which validates the method’s effectiveness.

Abstract The health of mobile software ecosystems is closely related to the interests of software developers, end‐users, and stakeholders. Therefore, it is crucial to maintain the mobile software ecosystem healthy and functioning. Researchers have done considerable research on mobile software ecosystems like Android and iOS. However, the evolution laws implicit in mobile software ecosystems have not attracted widespread attention. This paper proposes a research framework for investigating the evolution process and influencing factors of mobile software ecosystems based on community mining. Firstly, we mine the evolving ecosystem from many mobile software projects based on a community detection algorithm. Then we analyze the evolution process of the ecosystem by identifying evolution events in different periods. Furthermore, we utilize the multinomial logistics regression model to analyze the relevant indicators and summarize the crucial factors affecting the evolution. Meanwhile, by training the long short term memory (LSTM) model to predict evolution events, our prediction accuracy can reach 75%. This work can be used to maintain and improve the healthy operations of mobile software ecosystems.

Abstract Mobile application service ecosystems (MASEs) are highly complex systems that involve multiple factors influencing value creation. However, current research on the impact of structural holes (SHs) as an organizational characteristic on value creation in MASE is still insufficient. To address this research gap, this paper proposes a value creation model for MASE and investigates the impact of SH on the value creation of MASE. Moreover, this paper investigates the correlation between SH and diversity, which can facilitate regulating the value creation process in MASE. Finally, we construct a computational experiment, comparing and analyzing how changes in SH affect the value creation of MASE and the correlation between SH and diversity. The findings of this study can be used to induce the evolution of MASE and promote its value maximization.

Governance strategies related to platform ecosystems have become a vital issue for developing a smart society, attracting governments’ and practitioners’ attention. Under the consensus of “service as a commodity” and “platform as market,” service providers, platforms, services, and various supply demand matching methods form new supply processes. These elements are continuously and uncertainly changing during supply demand matching, which makes platform ecosystems constantly evolving. However, when multiple supply demand matching methods coexist such as service composition and crossover fusion, dynamic service supply and matching cause dilemmas in the platform ecosystem governance. To this end, this article proposes a model for platform ecosystem evolution with four dynamics: 1) dynamics between ISPs (services) and platforms; 2) dynamics between users and platforms; 3) dynamics among services; and 4) dynamics between services and demands. The model considers multiple supply demand matching methods and considers both fully online services and incompletely online services. Then, according to the market operation law, we design six evaluation indexes such as demand matching rate, service diversity, and market concentration to evaluate the efficiency of the platform market. Finally, a computational experiment system is established to simulate the dynamic supply and matching processes. The experimental results show that reducing the cost of service release can increase the amount of demand and the diversity of services, and the monopoly of digital platforms is a natural trend to improve the efficiency of supply and demand. The model provides a reference for the governance of platform ecosystems and lays a foundation for further research on the value cocreation mechanism of platform ecosystems.

Multi-scale detection based on Feature Pyramid Networks (FPN) has been a popular approach in object detection to improve accuracy. However, using multi-layer features in the decoder of FPN methods entails performing many convolution operations on high-resolution feature maps, which consumes significant computational resources. In this paper, we propose a novel perspective for FPN in which we directly use fused single-layer features for regression and classification. Our proposed model, You Only Look One Hourglass (YOLOH), fuses multiple feature maps into one feature map in the encoder. We then use dense connections and dilated residual blocks to expand the receptive field of the fused feature map. This output not only contains information from all the feature maps, but also has a multi-scale receptive field for detection. The experimental results on the COCO dataset demonstrate that YOLOH achieves higher accuracy and better run-time performance than established detector baselines, for instance, it achieves an average precision (AP) of 50.2 on a standard 3× training schedule and achieves 40.3 AP at a speed of 32 FPS on the ResNet-50 model. We anticipate that YOLOH can serve as a reference for researchers to design real-time detection in future studies. Our code is available at https://github.com/wsb853529465/YOLOH-main.

Heterogeneous information networks (HINs) are potent models of complex systems. In practice, many nodes in an HIN have their attributes unspecified, resulting in significant performance degradation for supervised and unsupervised representation learning. We developed an unsupervised heterogeneous graph contrastive learning approach for analyzing HINs with missing attributes (HGCA). HGCA adopts a contrastive learning strategy to unify attribute completion and representation learning in an unsupervised heterogeneous framework. To deal with a large number of missing attributes and the absence of labels in unsupervised scenarios, we proposed an augmented network to capture the semantic relations between nodes and attributes to achieve a fine-grained attribute completion. Extensive experiments on three large real-world HINs demonstrated the superiority of HGCA over several state-of-the-art methods. The results also showed that the complemented attributes by HGCA can improve the performance of existing HIN models.

Textbook question answering (TQA) task aims to infer answers for given questions from a multimodal context, including text and diagrams. The existing studies have aggregated intramodal semantics extracted from a single modality but have yet to capture the intermodal semantics between different modalities. A major challenge in learning intermodal semantics is maintaining lossless intramodal semantics while bridging the gap of semantics caused by heterogeneity. In this article, we propose an intermodal relation-aware heterogeneous graph network (IMR-HGN) to extract the intermodal semantics for TQA, which aggregates different modalities while learning features rather than representing them independently. First, we design a multidomain consistent representation (MDCR) to eliminate semantic gaps by capturing intermodal features while maintaining lossless intramodal semantics in multidomains. Furthermore, we present neighbor-based relation inpainting (NRI) to reduce semantic ambiguity via repairing fuzzy relations with correlations of relations. Finally, we propose hierarchical multisemantics aggregation (HMSA) to guarantee the completeness of semantics by aggregating features of nodes and relations with a reconstruction network (RN). Experimental results show that IMR-HGN could extract the intermodal semantics of answers, achieving a 2.16% improvement on the validation set of the TQA dataset and a 3.04% increase on the test set of the AI2D dataset.

Code search, which refers to the process of identifying the most relevant code snippets for a given natural language query, plays a crucial role in software maintenance. However, current approaches heavily rely on labeled data for training, which results in performance decreases when confronted with cross-domain scenarios including domain- or project-specific situations. This decline can be attributed to their limited ability to effectively capture the semantics associated with such scenarios. To tackle the aforementioned problem, we propose a ze R o-shot dom A in ada P tion with pre-tra I ned mo D els framework for code search named RAPID. The framework first generates synthetic data by pseudo labeling, then trains the CodeBERT with sampled synthetic data. To avoid the influence of noisy synthetic data and enhance the model performance, we propose a mixture sampling strategy to obtain hard negative samples during training. Specifically, the mixture sampling strategy considers both relevancy and diversity to select the data that are hard to be distinguished by the models. To validate the effectiveness of our approach in zero-shot settings, we conduct extensive experiments and find that RAPID outperforms the CoCoSoDa and UniXcoder model by an average of 15.7% and 10%, respectively, as measured by the MRR metric. When trained on full data, our approach results in an average improvement of 7.5% under the MRR metric using CodeBERT. We observe that as the model’s performance in zero-shot tasks improves, the impact of hard negatives diminishes. Our observation also indicates that fine-tuning CodeT5 for generating pseudo labels can enhance the performance of the code search model, and using only 100-shot samples can yield comparable results to the supervised baseline. Furthermore, we evaluate the effectiveness of RAPID in real-world code search tasks in three GitHub projects through both human and automated assessments. Our findings reveal RAPID exhibits superior performance, e.g., an average improvement of 18% under the MRR metric over the top-performing model.

Service recommendations are crucial for developers to create mashups such as mobile applications, workflows, e-business solutions, etc. Existing methods based on collaborative filtering or content analysis are manual and cannot automatically acquire services that align with the requirements of mashup creation. A possible solution to automatically acquiring necessary services for mashups is the seq2seq (sequence to sequence) generation model, which has demonstrated promising performance in automatic text and program code generation. However, two main challenges must be tackled in service acquisition based on the seq2seq model. First, the seq2seq model can only acquire a set of services without inter-service dependencies, but such dependencies are crucial in the generation of sequences for services. Second, external knowledge must be leveraged to recommend services more accurately that fulfill developers' requirements, such as similar historical user requirements and combining mashup category information, due to the incomplete description of user requirements. To tackle these challenges, this paper proposes GSR (Generation ofServiceRecommendations), an approach that can automatically acquire services based on user requirements. Specifically, GSR employs reinforcement learning to learn the inter-dependencies among services and integrate dependencies into service recommendations. To further improve the quality of the acquired services, GSR retrieves relevant user requirements based on BERT (Bidirectional Encoder Representation from Transformers) to help identify potential services. Experiment results conducted on real-world datasets show the superior performance of GSR. Compared with the existing recommendation approaches, the precision metric is increased by up to 1.99x, and the recall metric is increased by up to 12%.

Pre-trained language models (PLMs) have recently enabled rapid progress on sentiment classification under the pre-train and fine-tune paradigm, where the fine-tuning phase aims to transfer the factual knowledge learned by PLMs to sentiment classification. However, current fine-tuning methods ignore the risk that PLMs cause the problem of sentiment bias, that is, PLMs tend to inject positive or negative sentiment from the contextual information of certain entities (or aspects) into their word embeddings, leading them to establish spurious correlations with labels. In this paper, we propose an adaptive Gumbel-attacked classifier that immunes sentiment bias from an adversarial-attack perspective. Due to the complexity and diversity of sentiment bias, we construct multiple Gumbel-attack expert networks to generate various noises from mixed Gumbel distribution constrained by mutual information minimization, and design an adaptive training framework to synthesize complex noise by confidence-guided controlling the number of expert networks. Finally, we capture these noises that effectively simulate sentiment bias based on the feedback of the classifier, and then propose a multi-channel parameter updating algorithm to strengthen the classifier to recognize these noises by fusing the parameters between the classifier and each expert network. Experimental results illustrate that our method significantly reduced sentiment bias and improved the performance of sentiment classification.

Document-level Relation Extraction (DocRE)intends to extract relationships from documents.Some works introduce logic constraints into DocRE, addressing the issues of opacity and weak logic in original DocRE models.However, they only focus on forward logic constraints and the rules mined in these works often suffer from pseudo rules with high standardconfidence but low support.In this paper, we proposes Bidirectional Constraints of Beta Rules(BCBR), a novel logic constraint framework.BCBR first introduces a new rule miner which model rules by beta contribtion.Then forward and reverse logic constraints are constructed based on beta rules.Finally, BCBR reconstruct rule consistency loss by bidirectional constraints to regulate the output of the DocRE model.Experiments show that BCBR outperforms original DocRE models on relation extraction performance (∼2.7 F1) and logic consistency(∼3.1 Logic).Furthermore, BCBR consistently outperforms two other logic constraint frameworks.Our code is available at https://github.com/Louisliu1999/BCBR.

Graph Contrastive Learning (GCL) has recently enjoyed great success as an efficient self-supervised representation learning approach. However, the existing methods have focused on designing of contrastive modes and used data augmentation with a rigid and inefficient one-to-one sampling strategy. We adopted node neighborhoods to extend positive samplings and made avoided resorting to data augmentation to create different views. We also considered the homophily problem in Graph Neural Networks (GNNs) between the inter-class node pairs. The key novelty of our method hinged upon analyzing this GNNs problem and integrating the GCL sampling strategy with homophily discrimination, where we solved these two significant problems using one approach. We introduced a new parameterized neighbor sampling component to replace the conventional sub-optimal samplings. By keeping and updating the neighbor sets, both the positive sampling of GCL and the message passing of GNNs can be optimized. Moreover, we theoretically proved that the new method provided a lower bound of mutual information for unsupervised semantic learning, and it can also keep the lower bound with downstream tasks. In essence, our method is a new self-supervised approach, which we refer to as group discrimination, and it can make the downstream fine-tuning efficient. Our extensive empirical results demonstrate that the new method can significantly outperform the existing GCL methods because the former can solve the homophily problem in a self-supervised way with the new group discrimination method used.

Although large language models (LLMs) have demonstrated impressive performance in code generation, they still face challenges when dealing with complex code generation tasks. In the software development process, humans often refine complex tasks iteratively and continuously modify and improve them. Inspired by this, we propose a self-iteration code generation framework based on large language models like ChatGPT. To realize this idea, we incorporated software development methodologies into the self-iteration framework. We introduced four roles into each cycle, including analyst, designer, developer, and tester. Each role performs different tasks during the self-iteration cycle, with analyst and designer continuously improving requirements analysis and task design based on the testing feedback provided by tester, while developer are responsible for refactoring or modifying code until it passes testing, concluding the entire self-iteration process. We conducted extensive experiments on multiple benchmarks. The experimental results indicate: (1) The code generated by the self-iteration framework achieves up to a 21.3% relative improvement in Pass@1 compared to direct code generation. (2) The self-iteration framework also exhibits strong generalization performance, enhancing code generation quality for different large language models."Failure is simply the opportunity to begin again, this time more intelligently."- Henry Ford

Regarding code generation, researchers have recently proposed a retrieve-template-generation approach. This method involves retrieving similar code snippets through a retriever and providing them to a generator along with input descriptions. However, since the retrieved similar code can be influenced by various data types, it may lead the model to reference unrelated content, resulting in some discrepancies between the generated code and the target code. To mitigate this bias, we introduce a code generation method based on retrieve-template-generation called RTCoder. Specifically, RTCoder completes code generation through three steps: Retrieve. Using a natural language description, the retriever retrieves several similar code snippets from a corpus. Template. By comparing these similar code snippets, it employs the Rabin-Karp algorithm to extract their common substrings and represents different substrings with spaces, forming a code template. Generator. The generator, based on a specific natural language description and the corresponding code template, automatically generates the concrete target code. We conducted extensive comparative experiments on three datasets and used three widely used evaluation metrics. The experimental results demonstrate that: (1) Compared to mainstream code generation models, RTCoder shows improvements in all three metrics across different datasets. For instance, compared to the state-of-the-art CodeT5 base, the EM value is 5.98%, 3.34%, and 1.67% higher on the three datasets, respectively. (2) Our approach is effective for other models as well. Taking the CodeBLEU score on the Concode dataset as an example, the retrieval-template-based generation method improved by 3.73% and 1.80% compared to direct generation and retrieval-generation methods on the RNN model, respectively.

Neural networks are an emerging data-driven programming paradigm widely used in many areas. Unlike traditional software systems consisting of decomposable modules, a neural network is usually delivered as a monolithic package, raising challenges for some maintenance tasks such as model restructure and re-adaption. In this work, we propose DeepArc, a novel modularization method for neural networks, to reduce the cost of model maintenance tasks. Specifically, DeepArc decomposes a neural network into several consecutive modules, each of which encapsulates consecutive layers with similar semantics. The network modularization facilitates practical tasks such as refactoring the model to preserve existing features (e.g., model compression) and enhancing the model with new features (e.g., fitting new samples). The modularization and encapsulation allow us to restructure or retrain the model by only pruning and tuning a few localized neurons and layers. Our experiments show that (1) DeepArc can boost the runtime efficiency of the state-of-the-art model compression techniques by 14.8%; (2) compared to the traditional model retraining, DeepArc only needs to train less than 20% of the neurons on average to fit adversarial samples and repair under-performing models, leading to 32.85% faster training performance while achieving similar model prediction performance.

Social recommendations play a crucial role in providing personalized services to users by leveraging social relationships and user sessions. Despite recent advancements, it still faces challenges in dealing with social inconsistency and the loss of critical semantic information in user-service interactions. To overcome these problems, an Evolving Graph Contrastive Learning for Socially-aware Recommendation (EGCLSR) model is proposed for capturing users’ fresh interests. Specifically, the graph structure features on user-service interactions and the correlations between users and different sequences are extracted by the graph contrastive learning module. Then, social consistency sampling based on the graph convolutional network is adopted to filter out noise information effectively. Finally, time-sliced representations on the dual side (user, service) are integrated to capture users’ evolving interests by employing gated recurrent units. Comprehensive experiments on three datasets demonstrate the proposed model consistently outperforms the representative baseline methods in various evaluation metrics. EGCLSR facilitates the recommendation of services that fulfill instant requirements within dynamically evolving user interests.

The conventional crowdsourcing system is dependent on a centralized platform, which grants the platform owner undue authority to manipulate the operation of the system for unethical profits. In this paper, the crowdsourcing system is revolutionized in a decentralized manner with Blockchain as a Service (BaaS). All crowdsourcing operations are implemented with smart contracts deployed on blockchain. Requesters invoke these contracts to publish tasks, while workers invoke them to submit solutions. Notably, the operation of assigning tasks is regarded as a crowdsourcing task to be performed by assigners. Multiple assigners compute respective task assignment schemes in an off-chain manner, and then submit the schemes to blockchain for competition. Performance evaluations show that not only the operating efficiency of the crowdsourcing system is improved, but also the adverse consequences of the system being maliciously manipulated are avoided. The proposed decentralized crowdsourcing system is anticipated to restructure the business model of the conventional crowdsourcing industry.

Mobile Edge Computing (MEC) plays a crucial role in providing diverse computation and storage services to intelligent equipment. In recent years, the utilization of Unmanned Aerial Vehicles (UAVs) equipped with edge servers has emerged as a promising approach to enable ubiquitous edge computing services. This paradigm offers various benefits, such as reduced latency and flexible service provisioning. In the context of UAV-assisted edge computing, optimizing the location and trajectory of UAVs is vital, since the communication distance has significant impact on the communication rates between edge servers and devices. To address this optimization problem, this paper establishes a UAV-assisted edge computing system and focuses on investigating the path planning issue to expedite the offloading of computational tasks. In order to achieve this objective, we propose an improved tabu search algorithm that can efficiently optimize the number of UAVs, path planning. And to ensure reliable communication, we introduce reliability guarantees. Furthermore, we consider the energy limitations of UAVs to ensure practical feasibility. Through extensive simulations, we demonstrate that the proposed algorithm outperforms alternative approaches in terms of the specified objectives.

With the advent of mobile internet, Online-to-Offline (O2O) home services have emerged, such as home healthcare and repair services, greatly facilitating our lives. Customers book services through online platforms, and workers provide the requested services at the customers’ homes offline. However, each time workers travel to customers’ homes, they incur opportunity cost, leading to increased cost for home services. In this paper, we propose to bundle several O2O home service requests close to each other and match them with a worker. Therefore, requests that are in a bundle can split the worker’s opportunity cost. Specifically, we formalize the request bundling problem for O2O home services, which aims to minimize the overall cost of completing all requests while satisfying the time and Quality of Service(QoS) constraints. We present three Bi-layer Greedy request bundling(BiG) algorithms to solve it, including BiG-LEV, BiG-DIST, and BiG-COST. Besides, a cost accounting method based on Shapley value is designed to calculate the actual cost of each service for in-depth analysis. Finally, we illustrate a case of bundling requests for home healthcare services and compare the performance of the three algorithms.

Named entity recognition (NER) is the task of identifying and classifying named entities from texts. NER can benefit from linguistic dependency information, yet existing NER models can only utilize such information on datasets where dependency annotations are readily available. Dependency parsing (DP) models can be used to generate annotations, which are trained independent of the NER task and can cause error propagation to NER. In this paper, we propose a joint NER and DP model through multi-task learning, which allows the NER and DP modules to benefit from the joint training and provides an end-to-end solution to dependency-guided NER. Our model JOINDER uses a shared contextualized embedder, a word encoder, a biaffine dependency classifier, and a multi-hop dependency-guided NER. Experiments on several standard datasets in four languages show the effectiveness of joint learning and the outstanding performance of JOINDER compared to existing models. Moreover, our model can transfer dependency knowledge to other datasets with no dependency annotat.

Multi-interest recommendation methods extract multiple interest vectors to represent the user comprehensively. Despite their success in the matching stage, previous works overlook the long-tail problem. This results in the model excelling at suggesting head items, while the performance for tail items, which make up more than 70% of all items, remains suboptimal. Hence, enhancing the tail item recommendation capability holds great potential for improving the performance of the multi-interest model.

Smart contracts are self-executing programs on the blockchain that are critical to a range of industries, including finance, supply chain management, and healthcare.However, comprehending smart contracts can be challenging due to a lack of effective comments in most user-defined code.To address this challenge, we propose a novel retrieval-enhanced approach CC-GRA that leverages retrieval knowledge to generate high-quality comments for Solidity language code.Our approach carefully eliminates duplicated data and template data in the widely-used smart contract dataset to ensure a high-quality corpus.Extensive experiments and comprehensive analysis demonstrate the effectiveness applicability of our approach after being compared with eight state-of-the-art baselines.Finally, we conduct a human study and find the comment quality generated by our approach is better than baselines in terms of similarity, naturalness, and informativeness.

Abstract Named entity recognition (NER) is a fundamental part of other natural language processing tasks such as information retrieval, question answering systems and machine translation. Progress and success have already been achieved in research on the English NER systems. However, the Urdu NER system is still in its infancy due to the complexity and morphological richness of the Urdu language. Existing Urdu NER systems are highly dependent on manual feature engineering and word embedding to capture similarity. Their performance lags if the words are previously unknown or infrequent. The feature-based models suffer from complicated feature engineering and are often highly reliant on external resources. To overcome these limitations in this study, we present several deep neural approaches that automatically learn features from the data and eliminate manual feature engineering. Our extension involved convolutional neural network to extract character-level features and combine them with word embedding to handle out-of-vocabulary words. The study also presents a tweets dataset in Urdu, annotated manually for five named entity classes. The effectiveness of the deep learning approaches is demonstrated on four benchmarks datasets. The proposed method demonstrates notable progress upon current state-of-the-art NER approaches in Urdu. The results show an improvement of 6.26% in the F1 score.

Network embedding which is to learn a low dimensional representation of nodes in a network has been used in many network analysis tasks. Some network embedding methods, including those based on Generative Adversarial Networks (GAN) (a promising deep learning model), have been proposed recently. Existing GAN-based methods typically use GAN to learn a Gaussian distribution as a prior for network embedding, which makes it difficult to distinguish the node representation from Gaussian distribution. It did not apply the adversarial learning strategy on the representation mechanism but just on representation results. Thus, it does not make full use of the essential advantage of GAN, and leads to compromised performance of the method. To address this problem, we propose a novel adversarial learning framework consisting of three players for network embedding, which applies the adversarial learning strategy on the representation mechanism, called Adversarial representation mechanism GAN (ArmGAN). Specifically, the first two players, named encoder and competitor, aim to learn two different representation mechanisms (i.e., two ways projecting data onto latent space). They compete with each other to improve their representation mechanisms. The third player is the discriminator, which discriminate the representation mechanism of the encoder from that of the competitor. In addition, we design a perturbation strategy to produce fake networks from the original network, and feed the fake networks to the competitor to obtain a “fake” representation mechanism. We evaluated ArmGAN on a variety of tasks including node clustering, node classification, link prediction and visualization. Moreover, we compared ArmGAN with 10 state-of-the-art methods (including DGI, which is well-known for its high accuracy) on 7 real-world networks. The experimental results show the significant superiority of ArmGAN over the existing methods.

Recently, Action Recognition (AR) is facing the scalability problem, since collecting and annotating data for the ever-growing action categories is exhausting and inappropriate. As an alternative to AR, Zero-Shot Action Recognition (ZSAR) is getting more and more attention in the community, as they could utilize a shared semantic/attribute space to recognize novel categories without annotated data. Different from the AR focuses on learning the correlation between actions, ZSAR needs to consider the correlation of action-action, label-label and action-label at the same time. However, as far as we know, there is no work to provide structural guidance for the framework design of ZSAR according to its task characteristics. In this paper, we demonstrate the rationality of using the Energy-Based Model (EBM) to guide the framework design of ZSAR based on their inference mechanism. Furthermore, under the guidance of EBM, we propose an Energy-based Temporal Summarized Attentive Network (ETSAN) to achieve ZSAR. Specifically, to ensure the effectiveness of cross-modal matching, EBM needs to capture the correlations of input-input, output-output and input-output, based on discriminative and focused input and output space. To this end, we first design the Temporal Summarized Attentive Mechanism (TSAM) to capture the correlation of action-action by constructing discriminative and focused input space. Then, a Label Semantic Adaptive Mechanism (LSAM) is proposed to learn the correlation of label-label by adjusting the semantic structure according to the target task. Finally, we devise an Energy Score Estimation Mechanism (ESEM) to measure the compatibility (i.e. energy score) between video representation and label semantic embedding. With end-to-end training, our framework can capture all three of the correlations mentioned above simultaneously by minimizing the energy score of the correct action-label pair. Experiments on the HMDB51 and UCF101 datasets show that the proposed architecture achieves comparable results among methods based on the spatial-temporal visual feature of sequence-level, which demonstrates the efficiency of the EBM in guiding the framework design of ZSAR. In addition, our code is available athttps://github.com/oOHCIOo/ETSAN.

Software vulnerabilities, once disclosed, can be documented in vulnerability databases, which have great potential to advance vulnerability analysis and security research. People describe the key characteristics of software vulnerabilities in natural language mixed with domain-specific names and concepts. This textual nature poses a significant challenge for the automatic analysis of vulnerability knowledge embedded in text. Automatic extraction of key vulnerability aspects is highly desirable but demands significant effort to manually label data for model training. In this article, we propose unsupervised methods to label and extract important vulnerability concepts in textual vulnerability descriptions (TVDs). We focus on six types of phrase-based vulnerability concepts (vulnerability type, vulnerable component, root cause, attacker type, impact, and attack vector) as they are much more difficult to label and extract than name- or number-based entities (i.e., vendor, product, and version). Our approach is based on a key observation that the same-type of phrases, no matter how they differ in sentence structures and phrase expressions, usually share syntactically similar paths in the sentence parsing trees. Specifically, we present a source-target neural architecture that learns the Part-of-Speech (POS) tagging to identify a token’s functional role within TVDs, where the source neural model is trained to capture common features found in the TVD corpus, and the target model is trained to identify linguistically malformed words specific to the security domain. Our evaluation confirms that the proposed tagger outperforms (4.45%–5.98%) the taggers designed on natural language notions and identifies a broad set of TVDs and natural language contents. Then, based on the key observations, we propose two path representations (absolute paths and relative paths) and use an auto-encoder to encode such syntactic similarities. To address the discrete nature of our paths, we enhance the traditional Variational Auto-encoder (VAE) with Gumble-Max trick for categorical data distribution and thus create a Categorical VAE (CaVAE). In the latent space of absolute and relative paths, we further apply unsupervised clustering techniques to generate clusters of the same-type of concepts. Our evaluation confirms the effectiveness of our CaVAE, which achieves a small (85.85) log-likelihood for encoding path representations and the accuracy (83%–89%) of vulnerability concepts in the resulting clusters. The resulting clusters accurately label six types of vulnerability concepts from a TVD corpus in an unsupervised way. Furthermore, these labeled vulnerability concepts can be mapped back to the corresponding phrases in the original TVDs, which produce labels of six types of vulnerability concepts. The resulting labeled TVDs can be used to train concept extraction models for other TVD corpora. In this work, we present two concept extraction methods (concept classification and sequence labeling model) to demonstrate the utility of the unsupervisedly labeled concepts. Our study shows that models trained with our unsupervisedly labeled vulnerability concepts outperform (3.9%–5.14%) those trained with the two manually labeled TVD datasets from previous work due to the consistent boundary and typing by our unsupervised labeling method.

With the development of ICT (information and communications technology) and service economy, service ecosystem is emerging in a lot of fields, including E-commerce, O2O(Online To Offline) life service, healthcare service, cloud manufacturing, and so on. As a complex socio-technical system, the evolution of service ecosystem is the joint result of the interaction of the three heterogeneous networks, including social network, service network and value network. Under such circumstances, the traditional SOA (Service Oriented Architecture)-based analysis model is powerless. As a result, how to analyze the laws behind the evolution of service ecosystem is still a serious challenge in the field. This paper proposes a value oriented analysis framework (VOA) of service ecosystem, which can use value as a clue to describe the interaction of the three heterogeneous networks. In addition, a computational experiment system is established to verify the effectiveness of the VOA framework, which stimulates the effect of different intervention strategies on service ecosystem. The result shows that our analysis framework can provide new means and ideas for the analysis of service ecosystem.

Graph Convolutional Network (GCN) has shown remarkable potential of exploring graph representation. However, the GCN aggregating mechanism fails to generalize to networks with heterophily where most nodes have neighbors from different classes, which commonly exists in real-world networks. In order to make the propagation and aggregation mechanism of GCN suitable for both homophily and heterophily (or even their mixture), we introduce block modelling into the framework of GCN so that it can realize “block-guided classified aggregation”, and automatically learn the corresponding aggregation rules for neighbors of different classes. By incorporating block modelling into the aggregation process, GCN is able to automatically aggregate information from homophilic and heterophilic neighbors discriminately according to their homophily degree. We compared our algorithm with state-of-art methods which deal with the heterophily problem. Empirical results demonstrate the superiority of our new approach over existing methods in heterophilic datasets while maintaining a competitive performance in homophilic datasets.

Linjuan Wu, Shaojuan Wu, Xiaowang Zhang, Deyi Xiong, Shizhan Chen, Zhiqiang Zhuang, Zhiyong Feng. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Recommendation systems play a crucial part in helping users efficiently obtain information based on users’ current preferences and discover their individual needs, but the existing works are deficient in terms of the evolution of users’ interests. In this paper, Graph Embedding with Service and User information (GESU) model is proposed to address the limitations of capturing users’ fresh interests. Graph-structured data derived from time-varying session sequences are captured via gated graph neural networks. Then, the evolving influence of different services for users is obtained through a multi-head module. At the same time, a graph attention network is applied to predict users’ fresh consumption preferences by selecting representative friends to characterize user information. Extensive experiments on three datasets show that the proposed model outperforms state-of-the-art methods consistently on various evaluation metrics. GESU provides a means to recommend services that meet current requirements in an environment where users’ interests evolve dynamically.

The incompleteness of android documentation causes the lack of permission semantics, which in turn leads to permission misuse and threatens android system security. Previous studies have used API information to supplement permission semantics and recommended permissions, but they still lack sufficient contextual information, e.g. category and user review grades. To solve this problem, a novel Knowledge Graph based Convolutional Propagation Model (KGCP) is proposed for apps permission recommendation. In KGCP, we construct a knowledge graph (KG) to model the contextual information corresponding to apps and permissions. In order to regularize the representations of items, KGCP utilizes KG embedding technique to preserve its intrinsic structure while embedding entities and relationships into a continuous vector space. Focused on apps permission recommendation, KGCP learns the representation of entity through graph convolutional networks, which recursively aggregates information about neighbors in KG to mine potential preferences for app entities over permission entities. Experimental results show that KGCP improves by 16.7% over state-of-the-art apps permission recommendation methods, thus helps developers find the suitable permissions faster and more accurately.

The relation classification is to identify semantic relations between two entities in a given text. While existing models perform well for classifying inverse relations with large datasets, their performance is significantly reduced for few-shot learning. In this paper, we propose a function words adaptively enhanced attention framework (FAEA) for few-shot inverse relation classification, in which a hybrid attention model is designed to attend class-related function words based on meta-learning. As the involvement of function words brings in significant intra-class redundancy, an adaptive message passing mechanism is introduced to capture and transfer inter-class differences.We mathematically analyze the negative impact of function words from dot-product measurement, which explains why the message passing mechanism effectively reduces the impact. Our experimental results show that FAEA outperforms strong baselines, especially the inverse relation accuracy is improved by 14.33% under 1-shot setting in FewRel1.0.

Automatic function naming aims to generate a concise and meaningful name for a function, and has become a popular research area.Function naming models based on deeplearning have made significant progress in recent years.Most of the existing neural models represent a function based on the granularity of token or AST (Abstract Syntax Tree) node.However, generating function names requires more fine-grained knowledge of code, but the representation of tokens or AST nodes is not enough to capture global function semantics.In our work, we propose Apker, a novel Adaptive prior-knowledge-assisted function naming based on multi-level information explorer.The Apker includes three modules: Multi-level Information Explorer (MIE), Adaptive Prior Knowledge Adaptor (APKA) and the Generator.The MIE captures the function semantics from a local and global perspective, motivated by the understanding patterns of humans, who will first understand the meaning of each statement and then comb their logical relations to understand the whole function.The APKA uses the pre-retrieved prior knowledge to assist the model, motivated by our observation that certain name tokens can be extracted directly from certain statements and such probability differs significantly in different types of statements.Finally, the Generator generates function names.The experimental results demonstrate that our approach outperforms the baselines by 5.4% in Precision, 12.7% in Recall, and 7.4% in F1-score.

In recent years, as the demand for senior care services has further increased, it has become more difficult to obtain matching services from the vast amount of data. Therefore, this paper proposes a service recommendation framework PCE-CF based on an embedded user portrait model. The framework accurately describes the elderly users through four dimensions—population, society, consumption, and health—and constructs the user portrait model by embedding tags. The embedded vector of each older man is learned through the deep learning model, and different feature groups are meaningfully expressed in the transformation space. In addition, location context and dynamic interest model are introduced to process embedded vectors, and users' service preferences are predicted according to their dynamic behaviors. The experiment results show that the PCE-CF framework proposed in this paper can improve the recommendation algorithm's efficiency and have higher feasibility in personalized service recommendations.

Directed and undirected probabilistic graphical models have been successfully used in community detection in recent years, but existing graphical model based methods usually only use one type of probabilistic graphical model to discover communities. However, directed and undirected graphical models have their own advantages for characterizing different network information (attribute information and network topology). Intuitively, we can make use of the merit of both kinds of models by combining them into a unified model. However, combining directed and undirected graphical models is difficult, as they have different properties which prevent parameter sharing and joint training. In this article, we propose a unified model which integrates directed and undirected graphical models by transforming both them into factor graph. In addition, as network topology and attribute information may contain different degrees of noises, we add a selective attention layer to learn the reliable weight of each type of information source in node granularity. For training the model, we derive an iterative belief propagation algorithm to train all the parameters simultaneously. Extensive experiments on real networks and artificial benchmarks show the superiority of our approach over existing methods.

Discovering communities is an essential step in the analysis of complex systems, and it has two purposes: to identify functional modules and to interpret semantics. However, most of the existing community detection methods only focused on identify communities, while learning the semantics interpretation of communities has not been fully studied. In this paper, we focused on the problem of identifying communities and learning the semantics interpretation of modules jointly in an end-to-end model. We designed a novel generative model which combines two closely related parts, one for community discovery and the other for content clustering and semantics interpretation. By extracting the potential correlation between these two parts, our new method is not only robust to discovering communities, but also able to provide a community with more than one semantic topic. As for model inference, we developed a variational algorithm from a Bayesian point of view. Experimental results on the artificial benchmark and real networks showed the superior performance of the proposed approach over existing methods in terms of effectiveness and efficiency. We also analyzed semantic interpretability of community detection results through a case study over a large-scale music platform dataset.

With the development of cloud computing, service computing, IoT(Internet of Things) and mobile Internet, the diversity and sociality of services are increasingly apparent. With the increasing complexity of collaborative relationships between services, service ecosystems are beginning to emerge with the characteristics of natural ecosystems, economic systems and complex networks. Under this context, how to realize systematic evaluation of service ecosystem is of great significance to promote its sound development. Based on this, this article proposes a value entropy model that links the operating state of the system with the efficiency of value creation, which helps to clarify the performance of the service ecosystem from the perspective of multi-dimensional integration. In addition, a computational experiment system is established to verify the effectiveness of value entropy model, which stimulates the competitive evolution process of two service ecosystems with different strategies. The result shows that our model can provide new ideas for the analysis of service ecosystem evolution, and can also provide decision support for the optimization of operation strategy.

Trans-boundary has become not only a trend but also an important requirement and feature of Modern Service Industry. In Internet era, crossover service not only changes people's daily life, but also has a profound impact on the business model of traditional industry. However, because of the complexity and convergence of crossover service, it is difficult to analyze and identify whether crossover services are valuable and how to achieve them. In order to solve the problems, the analysis framework of crossover service is proposed from the perspective of value creation and realization, which includes four main parts: value assessment, data analysis, capability analysis, and execution path. Based on the proposed method, the New Retail in China is given as a case study to clarify the implementation process of crossover service. Furthermore, JD (self-employed mode) and Alibaba (platform mode) are taken as examples to explain how to migrate from different e-commerce models to the New Retail model. By introducing computational experiment, the performance of different execution paths can be evaluated. The comparison of experimental results with the reality shows that the proposed method can provide a roadmap for the implementation of crossover service.

Difficult samples of the minority class in imbalanced text classification are usually hard to be classified as they are embedded into an overlapping semantic region with the majority class.In this paper, we propose a Mutual Information constrained Semantically Oversampling framework (MISO) that can generate anchor instances to help the backbone network determine the re-embedding position of a non-overlapping representation for each difficult sample.MISO consists of (1) a semantic fusion module that learns entangled semantics among difficult and majority samples with an adaptive multi-head attention mechanism, (2) a mutual information loss that forces our model to learn new representations of entangled semantics in the non-overlapping region of the minority class, and (3) a coupled adversarial encoder-decoder that fine-tunes disentangled semantic representations to remain their correlations with the minority class, and then using these disentangled semantic representations to generate anchor instances for each difficult sample.Experiments on a variety of imbalanced text classification tasks demonstrate that anchor instances help classifiers achieve significant improvements over strong baselines.

With the rise of crossover services, service providers usually cooperate with each other on different platforms to expand their service value. However, in cross-platform cooperation, insufficient understanding and malicious competition between different platforms would lead to inaccurate trust establishment and unreliable trust recommendations. In this paper, we propose a trust management framework of cross-platform based on blockchain to establish, store and recommend trust securely for cross-platform cooperation. Firstly, we take into account the contextual background information to enhance interaction and understanding between platforms to achieve accurate trust establishment. Secondly, the trust recommendation algorithm is written into the blockchain in the form of smart contracts, which can ensure the security of trust recommendation. Finally, experiments are used to demonstrate the superiority and reliability of the framework.

Trust prediction is gaining significant interest since it could reduce the burden of user decision-makings effectively in various social activities. Existing works on trust prediction mainly based on trust networks, however, usually give little consideration to data sparsity and temporal continuity of user behavior. In order to solve these problems, we propose a comprehensive deep MemTrust model for trust prediction. With this model, we introduce a embedding layer to extend the feature space and alleviate the distinctive information oblivion caused by data sparsity. In addition, Long Short-Term Memory(LSTM) network is utilized to extract overall time series features through the multiple time slices of user features. Finally, the trust is estimated by pairwise time series features of users. Extensive experiments are validated on two real datasets, which demonstrate that the proposed model has superior performance compared with representative baseline approaches.

With the development of Online to Offline (O2O) model and the rapid growth of service types and numbers, service matching algorithms have become the key in connecting users and services. The traditional service matching algorithms lack consideration for the limited resources of O2O services, leading to the Matthew effect more seriously. In this context, how to alleviate the Matthew effect through the optimization of matching algorithms has become an urgent problem in this field. Based on this, this paper proposes an adaptive optimization algorithm of O2O service matching to achieve the balance of supply and demand by optimizing supply, thus alleviating the Matthew effect. In addition, a computational experiment system is constructed to verify the effect of different matching algorithms on alleviating the Matthew effect. The result shows that our proposed algorithm can provide new means and ideas for alleviating the Matthew effect.

The prosperous development of Internet services such as O2O, IoT, and Web API has brought new vitality to service commercial platforms. However, these services involve online and offline business, which are widely diverse without a unified design and development standard. In addition, Internet services update frequently, which leads to repeat releases on commercial platforms. Therefore, in this paper, we present a generic method to rapidly release Internet services on commercial platforms. The method uses a highly abstract metamodel to express service business extensively and realizes service functions by executing metamodel objects. This method has wide versatility. Meanwhile, it extends the DevOps theory to solve the frequent changes of service functions during use after the release. Finally, we verified the usability of this method in the elderly healthcare domain.

Aspect-based sentiment analysis (ABSA) aims to determine the sentiment polarity of the specific aspect for a given sentence. Attention-based models are widely used in this task because they can extract semantic information between context words to make up for the deficiency of sequence models in semantic encoding. In order to enhance the extraction of high-quality semantic information, we propose a novel joint model with Second-Order Features and Matching Attention (SOMA) for aspect-based sentiment analysis. Firstly, we introduce the second-order statistics to extract vital information and interact with the first-order features to generate the interaction representation. Secondly, we adopt Euclidean distance to replace the traditional matrix transformation to capture the semantic similarity between aspect terms and context words. Finally, we form a joint representation to focus on the meaningful words in the sentence. We conduct extensive experiments and comparisons on SemEval 2014, SemEval 2016, and Twitter datasets. Experimental results demonstrate the effectiveness of our model.

Query graph as a junction of semantic parsing in question answering over knowledge bases (KBQA) connects questions and logical queries. Though query graph consists of rich information such as structure, relation, etc., the current KBQA's models mainly utilize limited relation information in a naive way. It is not easy to learn the representation of a query graph with that information due to the heterogeneity of the query graph and intricate correlation of relations. In this paper, we propose a Global Semantic-based Message Passing (GSMP) model to model the global semantics of a query graph from its structure and relation information. In GSMP, we present a recurrent-based relational graph convolutional network (RGCN) to capture heterogeneous query graphs where the recurrent unit improves the capability of RGCN in processing small-scale query graphs. Moreover, we present a contextual-based method to remove ambiguity caused by intricate correlations where the contextual adjacency of relations optimizes relation representation. Finally, we present a nonlinear gate-based encoder to learning the representation of questions' syntactic tree, as the structure information of questions, for better matching the global semantics of query graphs. Experiments evaluated on benchmarks show that our model outperforms off-the-shelf models.

People usually describe the key characteristics of software vulnerabilities in natural language mixed with domain-specific names and concepts. This textual nature poses a significant challenge for the automatic analysis of vulnerabilities. Automatic extraction of key vulnerability aspects is highly desirable but demands significant effort to manually label data for model training. In this paper, we propose an unsupervised approach to label and extract important vulnerability concepts in textural vulnerability descriptions (TVDs). We focus on three types of phrase-based vulnerability concepts (root cause, attack vector, and impact) as they are much more difficult to label and extract than name- or number-based entities (i.e., vendor, product, and version). Our approach is based on a key observation that the same-type of phrases, no matter how they differ in sentence structures and phrase expressions, usually share syntactically similar paths in the sentence parsing trees. Therefore, we propose two path representations (absolute paths and relative paths) and use an auto-encoder to encode such syntactic similarities. To address the discrete nature of our paths, we enhance traditional Variational Auto-encoder (VAE) with Gumble-Max trick for categorical data distribution, and thus creates a Categorical VAE (CaVAE). In the latent space of absolute and relative paths, we further use FIt-TSNE and clustering techniques to generate clusters of the same-type of concepts. Our evaluation confirms the effectiveness of our CaVAE for encoding path representations and the accuracy of vulnerability concepts in the resulting clusters. In a concept classification task, our unsupervisedly labeled vulnerability concepts outperform the two manually labeled datasets from previous work.

In this paper, we study the problem of part-of-speech (POS) tagging for security vulnerability descriptions (SVD). In contrast to newswire articles, SVD often contains a high-level natural language description of the text composed of mixed language studded with codes, domain-specific jargon, vague language, and abbreviations. Moreover, training data dedicated to security vulnerability research is not widely available. Existing neural network-based POS tagging has often relied on manually annotated training data or applying natural language processing (NLP) techniques, suffering from two significant drawbacks. The former is extremely time-consuming and requires labor-intensive feature engineering and expertise. The latter is inadequate to identify linguistically-informed words specific to the SVD domain. In this paper, we propose an automatic approach to assign POS tags to tokens in SVD. Our approach uses the character-level representation to automatically extract orthographic features and unsupervised word embeddings to capture meaningful syntactic and semantic regularities from SVD. The character level representations are then concatenated with the word embedding as a combined feature, which is then learned and used to predict the POS tagging. To deal with the issue of the poor availability of annotated security vulnerability data, we implement a finetuning approach. Our approach provides public access to a POS annotated corpus of ~8M tokens, which serves as a training dataset in this domain. Our evaluation results show a significant improvement in accuracy (17.72%-28.22%) of POS tagging in SVD over the current approaches.

With the development of cloud computing, service computing, IoT(Internet of Things) and mobile Internet, the diversity and sociality of services are increasingly apparent. With the increasing complexity of collaborative relationships between services, service ecosystems are beginning to emerge with the characteristics of natural ecosystems, economic systems and complex networks. Under this context, how to realize systematic evaluation of service ecosystem is of great significance to promote its sound development. As shown in Fig.1 , the value creation of service ecosystem consists of three elements: Input, Output, and Operation. Input means customers’ value demands, which drives the constant evolution of service ecosystem. Output means the value created by the service ecosystem in a certain period of time. Operation means the value creation ability of service ecosystem.

Summary Web services (or Web APIs) on the Internet tends to encounter various unexpected runtime failures because of their dynamicity and distribution. Self‐adaptation technologies for the service‐based business process can effectively repair runtime failures and improve its success rate. However, the same failures may occur on subsequent invocations because relevant processes do not evolve after failures. This makes the response time of the business processes too long. We proposed a self‐adaptation and distributed knowledge‐based evolution model (SDKEM) to guarantee business processes' stabilities, that is, low failure rates and stable response time. SDKEM adopts a service knowledge base (SKB) to organize services from a provider and uses bridge rules to eliminate semantic conflicts among multiple distributed SKBs. It can automatically trigger the evolution of a service ecosystem through the designed self‐adaptation mechanism. We adopt the “survival of the fittest” principle for crucial elements in the ecosystem during evolution so that ultimately, service‐based processes and services with high stability remain. Experiments show that, with the developed evolution mechanism, runtime failures of business processes significantly reduce. In most cases, their response time and success rates are comparable to those under the running situation where no runtime failure occurs, meaning the runtime failures within a service‐based process are automatically repaired.

Abstract Ontology-mediated querying (OMQ) provides a paradigm for query answering according to which users not only query records at the database but also query implicit information inferred from ontology. A key challenge in OMQ is that the implicit information may be infinite, which cannot be stored at the database and queried by off -the -shelf query engine. The commonly adopted technique to deal with infinite entailments is query rewriting, which, however, comes at the cost of query rewriting at runtime. In this work, the partial materialization method is proposed to ensure that the extension is always finite. The partial materialization technology does not rewrite query but instead computes partial consequences entailed by ontology before the online query. Besides, a query analysis algorithm is designed to ensure the completeness of querying rooted and Boolean conjunctive queries over partial materialization. We also soundly and incompletely expand our method to support highly expressive ontology language, OWL 2 DL. Finally, we further optimize the materialization efficiency by role rewriting algorithm and implement our approach as a prototype system SUMA by integrating off-the-shelf efficient SPARQL query engine. The experiments show that SUMA is complete on each test ontology and each test query, which is the same as Pellet and outperforms PAGOdA. Besides, SUMA is highly scalable on large datasets.

Imbalanced text classification, as practical and essential text classification, is the task to learn labels or categories for imbalanced text data. Existing imbalanced text classification approaches are mostly based on the Imbalance Ratio (i.e. ratio of sizes between categories). Recently, some researchers verified that the imbalance ratio severely affects the performance of classifiers when intrinsic characteristics of data such as class overlapping and small disjuncts occur. However, since the distribution of real-world data is unknown, it is difficult to describe above intrinsic characteristics directly. In this paper, we transform the unknown distribution of data into a graph model and present a graph-based imbalance index named GIR to predict the impact of imbalanced text data on classification performance. Firstly, we introduce an environmental factor that makes the imbalance index sensitive to the intrinsic characteristics of data. Secondly, we propose a graph-based method to calculate this environmental factor. Finally, we use the imbalance index to analyze the performances of imbalanced learning methods and the impact of imbalanced data on text classifiers. The experimental results evaluated on both synthetic data sets and real-world data sets demonstrate the effectiveness of our approach.

To battle the ever-increasing Android malware, malware family classification, which classifies malware with common features into a malware family, has been proposed as an effective malware analysis method. Several machine-learning based approaches have been proposed for the task of malware family classification. Our study shows that malware families suffer from several data imbalance, with many families with only a small number of malware applications (referred to as few shot malware families in this work). Unfortunately, this issue has been overlooked in existing approaches. Although existing approaches achieve high classification performance at the overall level and for large malware families, our experiments show that they suffer from poor performance and generalizability for few shot malware families, and traditionally downsampling method cannot solve the problem. To address the challenge in few shot malware family classification, we propose a novel siamese-network based learning method, which allows us to train an effective MultiLayer Perceptron (MLP) network for embedding malware applications into a real-valued, continuous vector space by contrasting the malware applications from the same or different families. In the embedding space, the performance of malware family classification can be significantly improved for all scales of malware families, especially for few shot malware families, which also leads to the significant performance improvement at the overall level.

Emotional intelligence is a crucial part for human-machine dialogue system. However, the existing research on dialogue mainly faces three problems: (1) focus on the content level of each response while ignoring the impact of emotional factors in the multi-turn dialogue; (2) lacking scalability and adaptability is that only the emotion categories specified by users are generated in a single-turn dialogue; (3) it is difficult to capture and perceive fine-grained emotions and the speaker's emotional state according to the emotional context. To address these problems, we propose an emotional expression model in multi-turn dialogue (EmoEM), which combines emotion-semantic graph with multitask learning mechanism, applying the dialogue generator based on seq2seq network and graph convolution network (GCN) to generate more natural and personalized emotional responses in a structured manner. Generally, EmoEM considers constructing emotion-semantic graph to describe explicit and implicit emotions dynamically. Then, the emotion-semantic graph is applied to the dialogue generator based on the seq2seq neural network, mainly to improve the semantic consistency and text quality in multi-turn dialogue. Moreover, multi-task learning mechanism is introduced to enhance the emotional expression of the text and obtain expected emotional responses. The experimental results show that EmoEM outperforms several baselines in BLEU, diversity and emotional expression.

Community detection, aiming at partitioning a network into multiple substructures, is practically importance. Graph convolutional network (GCN), a new deep-learning technique, has recently been developed for community detection. Markov Random Fields (MRF) has been combined with GCN in the MRFasGCN method to improve accuracy. However, the existing GCN community-finding methods are semi-supervised, even though community finding is essentially an unsupervised learning problem. We developed a new GCN approach for unsupervised community detection under the framework of Autoencoder. We cast MRFasGCN as an encoder and then derived node community membership in the hidden layer of the encoder. We introduced a community-centric dual decoder to reconstruct network structures and node attributes separately in an unsupervised fashion, for faithful community detection in the input space. We designed a scheme of local enhancement to accommodate nodes to have more common neighbors and similar attributes with similar community memberships. Experimental results on real networks showed that our new method outperformed the best existing methods, showing the effectiveness of the novel decoding mechanism for generating links and attributes together over the commonly used methods for reconstructing links alone.

Complex networks are widely used in the research of social and biological fields. Analyzing real community structure in networks is the key to the study of complex networks. Modularity optimization is one of the most popular techniques in community detection. However, due to its greedy characteristic, it leads to a large number of incorrect partitions and more communities than in reality. Existing methods use the modularity as a Hamiltonian at the finite temperature to solve the above problem. Nevertheless, modularity is not formalized as a statistical model in the method, which makes many statistical inference methods limited and cannot be used. Moreover, the method uses the sum-product version of belief propagation (BP) and its performance is not as good as the max-sum version, since it calculates per-variable marginal probabilities rather than the joint probability . To address these issues, we propose a novel Markov Random Field (MRF) method by formalizing modularity as an energy function based on the rich structures of MRF to represent properties and constraints of this problem, and use the max-sum BP to infer model parameters. In order to analyze our method and compare it with existing methods, we conducted experiments on both real-world and synthetic networks with ground-truth of communities, showing that the new method outperforms the state-of-the-art methods.

Although Deep Convolutional Neural Networks (DCNNs) facilitate the evolution of 3D human pose estimation, ambiguity remains the most challenging problem in such tasks. Inspired by the Human Perception Mechanism (HPM), we propose an image-to-pose coding method to fill the gap between image cues and 3D poses, thereby alleviating the ambiguity of 3D human pose estimation. First, in 3D pose space, we divide the whole 3D pose space into multiple subregions named pose codes , turning a disambiguation problem into a classification problem. The proposed coding mechanism covers multiple camera views and provides a complete description for 3D pose space. Second, it is noteworthy that the articulated structure of the human body lies on a sophisticated product manifold and the error accumulation in the chain structure will undoubtedly affect the coding performance. Therefore, in image space, we extract the image cues from independent local image patches rather than the whole image. The mapping relationship between image cues and 3D pose codes is established by a set of DCNNs. The image-to-pose coding method transforms the implicit image cues into explicit constraints. Finally, the image-to-pose coding method is integrated into a linear matching mechanism to construct a 3D pose estimation method that effectively alleviates the ambiguity. We conduct extensive experiments on widely used public benchmarks. The experimental results show that our method effectively alleviates the ambiguity in 3D pose recovery and is robust to the variations of view.

With the increasing popularity and great economic benefit from vulnerability exploitation, it is important to study mobile vulnerability in the mobile ecosystem. Beyond the traditional technical solutions such as developing technologies to identify potential vulnerabilities, discover the widely available exploitations and protect consumers from attacks, constructing the vulnerability market, a marketplace for vulnerability discovery, disclosure and exploitation, has been considered as an effective approach. Therefore, understanding the mechanism of the vulnerability market for further optimizations is attracting attentions from both academia and industry. Since mobile ecosystem is playing an increasingly important role for the daily life, this paper aims to understand the evolution of the mobile vulnerability market in a data-driven approach, aiming to identify the important issues for further research. Specially, a five-layer heterogeneous network, consisting of the software vendors, products, public disclosed vulnerabilities, hunters, organizations and their relations, is established to formally represent the evolution of the mobile vulnerability market. Based on the data collected from a variety of agencies, including NVD, OSVDB, BID and vendor advisories, a comprehensive empirical analysis is reported, focusing on the growth of the mobile vulnerability market as well as the interactions between mobile and other PCs platforms. Finally, suggestions drawn from the observations, including security evaluation for code reused, data leaking protection and permission overuse identification, hunter's strategy and behavior understanding, information sharing and external workforce hiring, as well as cross-platform vulnerability digging are discussed for further security enhancement.

In this paper, we lead the first efforts towards intelligent RDF data management in SSDs. We propose to deeply fuse the RDF data in SSDs. In detail, the operations (e.g., data query) applied to RDF can be directly achieved in SSDs. To this end, we explore two RDF data organizations (e.g., triple-based) with the consideration of the internal structure of SSDs. The experiment is conducted on the Patient Disease Drug (PDD) Graph dataset [11]. The experimental results show that the proposed two strategies achieve the comprehensive, scalable in-SSD computation from different aspects (e.g., space efficiency or query efficiency).

Most covariance-based representations of actions are focused on the statistical features of poses by empirical averaging weighting. Note that these poses have a variety of saliency levels for different actions. Neglecting pose saliency could degrade the discriminative power of the covariance features, and further reduce the performance of action recognition. In this paper, we propose a novel saliency weighting covariance feature representation, Saliency-Pose-Attention Covariance(SPA-Cov), which reduces the negative effects from the ambiguous pose samples. Specifically, we utilize a discriminative approach to derive probability distribution of action categories for each pose, which is modeled by the uncertainty of information entropy to obtain the salient weighting. Experimental results show that our proposed method efficiently improves the discriminative power of the generated covariance. In some databases, the proposed SPA-Cov outperforms the state-of-the-art variant methods which are based on kernel matrix, Bayesian posterior features, temporal hierarchical features, etc.

Software vulnerabilities seriously affect the security of computing systems and they are continuously disclosed and reported. When documenting software vulnerabilities, characterizing the severity, exploitability and impact of a vulnerability is critical for effective triaging and management of software vulnerabilities. Faced with ever-growing number of new vulnerabilities, we observe a significant lag between the disclosure of a vulnerability and the specification of its characteristics. This lag calls for automated, reliable assessment of vulnerability characteristics to assist security analysts in allocating their limited efforts to potentially most serious vulnerabilities. Existing automated techniques for vulnerability assessment require hand-crafted features and balanced data, and consider each specific characteristic independently at a time. In this paper, we propose a multi-task machine learning approach for the joint prediction of multiple vulnerability characteristics based on the vulnerability descriptions. Our approach gets rid of the requirement of balanced data, and it relies on neural networks that learn to extract features from training data. Using the large-scale vulnerability data in the Common Vulnerabilities and Exposure(CVE) database, we conduct extensive experiments to compare different configurations of neural network feature extractors, study the impact of multi-task learning versus independent-task learning, and investigate the performance of our approach for predicting the characteristics of newly disclosed vulnerabilities and the minimum requirement of historical vulnerability data for training reliable prediction model.

Individual identification from skeleton sequence is a challenging task because various covariate factors may produce drastic changes in human poses, thus causing large intra-class variability. In this paper, we present a novel spatio-temporal multi-factor discriminant analysis (ST-MFDA) to reduce the impact of covariate factors on identification performance. In our approach, a set of paired factor-specific spatio-temporal projections are learned to project motion features from different factors into a common discriminant subspace. In the common subspace, features from the same individual are united and those from different individuals are separated by Fisher discriminant criterion. We show that spatio-temporal projections of various factors can be jointly learned by solving an iterative generalized eigenvalue problem. According to the experimental results from three public databases, we confirm that spatio-temporal projections trained in this way lead to significant improvements in identification accuracy under the different covariate factors.

Crossover cooperation and fusion between services is becoming very common in the modern service industry. Crossover service fusion can create value that cannot be provided by single-domain services, thus achieving the value-emergence effect of 1+1>2. However, semantic inconsistencies in business and interface make crossover service fusion difficult and time-consuming. This paper proposes an interactive crossover service fusion approach based on microservice architecture to enable smooth and rapid integration of domain services. This approach takes service fusion requirements as the driving force to detect business inconsistencies between the services to be fused, and carries out business process reengineering by human-computer interaction. Then, semantic inconsistencies in service interface matching are detected and solved by splitting and completing parameter concepts to obtain the service fusion design scheme. Finally, the implementation scheme based on microservice architecture transforms the business coupling between domain services into asynchronous data communication, which facilitates the crossover fusion of complex business services. The elderly healthcare application is used to demonstrate and validate our approach.

In the context of big data, cloud storage services provide users with the ideal data storage service. But external store of data causes cloud storage service providers gain control of the data. Therefore, our work should consider how to ensure the privacy of data and maintain the integrity of data when enjoying convenient services. This paper builds a blockchain-based solution through research on cloud storage service model and blockchain technology. And related protocols are built on the solution-based architecture. In our solution, the decentralized model solves the single point of trust problem in the traditional data auditing service model by collective trust. A public agreement enables auditors to efficiently build proof of data integrity without touching data. The protocol allows users to trace the history of their data, and examine whether the owner of the data guarantees the privacy of the data in an after-the-fact audit. In addition, our work constructed the prototype system of the scheme and revealed the effectiveness of the scheme through system testing.

Mobile edge computing has significantly reduced the response time of mobile applications by executing services in close proximity to mobile consumers. However, the capabilities and resources of edge servers are typically limited; additionally, service requirements in mobile environments are becoming increasingly complex and diverse. In this context, properly dispatching service requests to edge and cloud servers to improve the quality of services has become a crucial problem. In this paper, we focus on this problem and aim to minimize the response time of service invocations in mobile edge computing systems. The problem is formulated as an optimization problem, and a heuristic algorithm that combines the Genetic algorithm and the simulated Annealing algorithm for service selection in Mobile Edge Computing systems (GAMEC) is proposed to solve the problem. A series of experiments has demonstrated that the method performs well in reducing the response time of service invocations in mobile edge computing systems. Moreover, the execution time of GAMEC is of a low order of magnitude, and the algorithm scales well as the experimental scale increases.

Markov Random Field (MRF) has been successfully used in community detection recently. However, existing MRF methods only utilize the network topology while ignore the semantic attributes. A straightforward way to combine the two types of information is that, one can first use a topic clustering model (e.g. LDA) to derive group membership of nodes by using the semantic attributes, then take this result as a prior to define the MRF model. In this way, however, the parameters of the two models cannot be adjusted by each other, preventing it from really realizing the complementation of the advantages of the two. This paper integrates LDA into MRF to form an end-to-end learning system where their parameters can be trained jointly. However, LDA is a directed graphic model whereas MRF is undirected, making their integration a challenge. To handle this problem, we first transform LDA and MRF into a unified factor graph framework, allowing sharing the parameters of the two models. We then derive an efficient belief propagation algorithm to train their parameters simultaneously, enabling our approach to take advantage of the strength of both LDA and MRF. Empirical results show that our approach compares favorably with the state-of-the-art methods.

AI benchmarking becomes an increasingly important task. As suggested by many researchers, Intelligence Quotient (IQ) tests, which is widely regarded as one of the predominant benchmarks for measuring human intelligence, raises an interesting challenge for AI systems. For better solving IQ tests automatedly by machines, one needs to use, combine and advance many areas in AI including knowledge representation and reasoning, machine learning, natural language processing and image understanding. Also, automated IQ tests provides an ideal testbed for integrating symbolic and sub-symbolic approaches as both are found useful here. Hence, we argue that IQ tests, although not suitable for testing machine intelligence, provides an excellent benchmark for the current development of AI research. Nevertheless, most existing IQ test datasets are not comprehensive enough for this purpose. As a result, the conclusions obtained are not representative. To address this issue, we create IQ10k, a large-scale dataset that contains more than 10,000 IQ test questions. We also conduct a comparison study on IQ10k with a number of state-of-the-art approaches.

In this paper, we present a neural network (InteractionNN) for sparse predictive analysis where hidden features of sparse data can be learned by multilevel feature interaction. To characterize multilevel interaction of features, InteractionNN consists of three modules, namely, nonlinear interaction pooling, layer-lossing, and embedding. Nonlinear interaction pooling (NI pooling) is a hierarchical structure and, by shortcut connection, constructs low-level feature interactions from basic dense features to elementary features. Layer-lossing is a feed-forward neural network where high-level feature interactions can be learned from low-level feature interactions via correlation of all layers with target. Moreover, embedding is to extract basic dense features from sparse features of data which can help in reducing our proposed model computational complex. Finally, our experiment evaluates on the two benchmark datasets and the experimental results show that InteractionNN performs better than most of state-of-the-art models in sparse regression.

Cross-project defect prediction (CPDP) as a means to focus quality assurance of software projects was under heavy investigation in recent years. In this paper, we propose a novel CPDP approach via deep learning. In particular, we model each program module via simplified abstract syntax tree (S-AST). For each node in S-AST, only the project-independent node type is remained and other project-specific information (such as name of variable and method) is ignored, so that the modeling method is project-independent and suitable for CPDP issue. Then we extract token sequences from program modules modeled as S-AST. In addition, to construct meaningful vector representations for token sequences, we propose a novel unsupervised embedding method ASTToken2Vec, which learns semantic information from S-AST's natural structure. Finally, we use BLSTM (bi-directional long short-term memory) based neural network to automatically learn semantic features from vectorized token sequences and construct CPDP models. In our empirical studies, 10 real large-scale open source Java projects are chosen as our empirical subjects. Final results show that our proposed CPDP approach can perform significantly better than 5 state-of-the-art CPDP baselines in terms of AUC.

In this paper, we propose a holistic solution to address several important and challenging issues in storage data management in light of emerging NVDIMM-based architecture: namely, new performance modeling, NVDIMM-based migration, and architectural support for NVDIMMs on migration optimization. In particular, a novel NVDIMM-based heterogeneous storage performance model is proposed to effectively address bus contention issues caused by placing NVDIMMs on the memory bus. We also develop an NVDIMM-based lazy migration scheme to effectively minimize adverse effects caused by memory traffic interferences during storage data management processes. Finally, the NVDIMM-based architectural support for migration optimization is proposed to increase channel parallelism in the destination NVDIMMs and bypass buffer caches in the source NVDIMMs, so that the impact of memory traffic can be alleviated. We present detailed evaluation and analysis to quantify how well our techniques can enhance the I/O performances of big workloads via efficient heterogeneous storage hierarchy management. Our experimental results show that overall the proposed techniques yield up to 98% performance improvement over the state-of-the-art techniques.

Summary form only given, as follows. The complete presentation was not made available for publication as part of the conference proceedings. Aiming at community and home based aging care on Internet, an integration Platform of Health Services in Aging care is developed based on the intelligent sensing, physical transmission, O2O Services, cloud computing and big data technology. Intelligent home and health monitoring equipment detect and record the living environment, vital signs and physiological parameters, daily behavior characteristics and recessive psychological and emotional data of the elderly. These data are upload to the cloud to establish health files for the elderly. The elderly and their children can know their individual health status and overall development trend at any time, so as to predict, analyze and formulate life care, safety guardianship, rehabilitation and health care and medical treatment programs. Relying on this service platform, we focus on providing a solution combining software and hardware for community and home-based pension management services to meet the needs of the elderly community-at-home healthy pension in all respects.

To guarantee the quality of software, specifying security requirements (SRs) is essential for developing systems, especially for security-critical software systems. However, using security threat to determine detailed SR is quite difficult according to Common Criteria (CC), which is too confusing and technical for non-security specialists. In this paper, we propose a Co-occurrence Recommend Model (CoRM) to automatically recommend software SRs. In this model, the security threats of product are extracted from security target documents of software, in which the related security requirements are tagged. In order to establish relationships between software security threat and security requirement, semantic similarities between different security threat is calculated by Skip-thoughts Model. To evaluate our CoRM model, over 1000 security target documents of 9 types software products are exploited. The results suggest that building a CoRM model via semantic similarity is feasible and reliable.

In an ultra-dense fifth-generation (5G) wireless network, the deployment of dense cells with a large number of base stations can cause frequent handovers, which could lead to higher handover delays and some security issues to make the network difficult to meet various requirements of 5G wireless networks. To target this problem, we propose to employ Mobile Edge Computing (MEC) servers into the traditional authentication architecture for re-authentication. Moreover, we improve the existing Extensible Authentication Protocol-Authentication and Key Agreement (EAP-AKA) protocol to design a new authentication mechanism over this authentication architecture. To enhance security functionality, pseudonyms are used instead of permanent names and a one-way hash function is used for the authentication. The results of formal verification and performance evaluation show that the proposed handover authentication mechanism can reduce the authentication delays and ensuring the security of the proposed handover authentication scheme.

As a new form of manufacturing industry in the Internet era, social manufacturing has its inherent “social-cyber” complexity: the source of manufacturing service is social, and such sociality aggravates the diversity, uncertainty, and dynamics of service supply. This poses new challenges to the service matching between supply-side and demand-side. In order to meet this challenge, it is necessary to conduct a complexity analysis of social manufacturing. Traditional researches mainly rely on data statistics and macro analysis, in which there are difficulties in clearly identifying the links between various impact factors and macro evolution phenomena. In order to change such a situation, this paper proposes a modeling framework of social manufacturing from the aspect of social learning evolution (SLE), including individual evolution model, organizational learning model, and social learning model. Based on the SLE framework, the corresponding computational experiment system is built to analyze the complexity of social manufacturing. The performance of several evolution mechanisms in social manufacturing is simulated and compared as a case study to present the application of SLE framework. The results demonstrate that our method has a substantial promise.

With the sweeping progress of service computing technology and crowdsourcing, individuals are offering their capability as human services online. Companies are orchestrating human services for complex problem-solving, resulting in the rapid growth of human service ecosystems nowadays. Considering the unique characteristics of human services, like capability growth and human-involving collaboration, it is essential to understand the patterns of the development and collaboration among human services. Therefore, this paper proposes a three-layer time-aware heterogeneous network model to quantify the evolution in the human service ecosystem. Based on the model, an exploratory empirical study is presented to uncover how human service providers and consumers develop their capability in service provision and orchestration, as well as how human services collaborate with each other over time. Insights from the emerging patterns open a gateway for further research to facilitate human service adoption, including human service composition recommendation, human skill expansion suggestion, and systematic mechanism design.

Semantic annotations play an important role in semantics-aware service discovery, recommendation and composition. While existing approaches and tools focus on facilitating the development of semantic annotations on web services, the validation of the quality of annotations is largely overlooked. Meanwhile, the refinement of semantic annotations mostly goes through manual processes, which not only is time-consuming but also requires significant domain knowledge. To enhance the Quality of Semantic Annotation (QoSA), we have developed a technique to incrementally assess and correct semantic annotations of web services. Aiming at supporting web service interoperation, we have formalized the QoSA of input and output parameters. Based on such formalism, test cases are automatically generated to validate service annotations. Learned semantic instances are then accumulated to iteratively validate semantic annotations of other services. Furthermore, a three-phase optimization methodology including local-feedback, global-feedback, and global-propagate is developed to improve the QoSA by incrementally correcting inaccurate annotations. Experiments over a real-world web services repository have demonstrated that our technique can effectively improve QoSA of services, gaining a 78.68 percent improvement in input parameters annotations and identifying 36.47 percent inaccurate output parameters annotations. The proposed technique can be equipped at various service repositories to enhance service discovery and recommendation.

Markov Random Field (MRF) is a powerful framework for developing probabilistic models of complex problems. MRF models possess rich structures to represent properties and constraints of a problem. It has been successful on many application problems, particularly those of computer vision and image processing, where data are structured, e.g., pixels are organized on grids. The problem of identifying communities in networks, which is essential for network analysis, is in principle analogous to finding objects in images. It is surprising that MRF has not yet been explored for network community detection. It is challenging to apply MRF to network analysis problems where data are organized on graphs with irregular structures. Here we present a network-specific MRF approach to community detection. The new method effectively encodes the structural properties of an irregular network in an energy function (the core of an MRF model) so that the minimization of the function gives rise to the best community structures. We analyzed the new MRF-based method on several synthetic benchmarks and real-world networks, showing its superior performance over the state-of-the-art methods for community identification.

The major challenge for individual identification from periodical locomotion is to explore the unique spatio-temporal motion characteristic of each individual. In this paper, we present a novel spatio-temporal metric learning approach to activate the discriminating components of Riemannian features. Specifically, to deal with articulated motion embedded in a high dimensional space, we extend the Active Appearance Model to Riemannian manifold to measure the intrinsic variations of poses. Then we extract high order geometric features of each joint, which naturally suggests the biometric signatures of enrolled individuals. To model the most discriminant features in a linear subspace, we propose a Spatio-Temporal Large Margin Nearest Neighbor (ST-LMNN) algorithm to learn the low-dimensional linear embedding in the spatial and temporal domain, respectively. According to the experimental results from two public databases and a new built database, the proposed approach can achieve more accurate identification results in walking and running.

In this paper, we present an embedding-based framework (TrQuery) for recommending solutions of a SPARQL query, including approximate solutions when exact querying solutions are not available due to incompleteness or inconsistencies of real-world RDF data. Within this framework, embedding is applied to score solutions together with edit distance so that we could obtain more fine-grained recommendations than those recommendations via edit distance. For instance, graphs of two querying solutions with a similar structure can be distinguished in our proposed framework while the edit distance depending on structural difference becomes unable. To this end, we propose a novel score model built on vector space generated in embedding system to compute the similarity between an approximate subgraph matching and a whole graph matching. Finally, we evaluate our approach on large RDF datasets DBpedia and YAGO, and experimental results show that TrQuery exhibits an excellent behavior in terms of both effectiveness and efficiency.

Service clustering is the foundation of service discovery, recommendation and composition. Most of the existing methods mainly use service attribute information and ignore the semantic-based invocation relationships among service users. In fact, mutual invocation relationships between services occur on operations of the corresponding services, while service attributes are the whole service description. Our main challenge may be to effectively combine these two kinds of data for service clustering. To address this issue, we propose a new probabilistic generative model which contains two closely connected parts, one characterizing operation community memberships by using operation invocation relationships, and the other characterizing service cluster memberships by utilizing service attributes. The correlations between these two parts are characterized by the relationships between operation communities and service clusters. To train this model, we provide a nested expectation-maximization algorithm. Experimental results show its superior performance over the existing methods for service clustering.

Semantic Annotation plays an essential role in automatic service discovery and composition. However, existing approaches and tools cannot achieve high annotation quality to ensure the semantic service application. Meanwhile, the semi-automatic strategies for improving the annotation quality are time-consuming. To further improve the efficiency as well as the quality of the annotation, this paper presents an effective method involving human-computer interaction to further optimize the annotation procedure. Besides employing the feedback and propagation strategy to semi-automatically improve the annotation quality, the strategy to involve the manual annotation is developed when the efficiency of semi-automatically strategy is related low. To optimize the manual annotation procedure, a clustering based approach is presented to select the most impacted candidates to optimize the annotation improvement. In addition, to help the annotators to choose the correct annotation, the local ontology restriction based method is further designed to improve the recommendation performance. The experiments show that our approach effectively involving the human intervention can significantly improve the annotation quality, faster the quality improvement procedure and reduce the manual load by increasing the recommendation accuracy.

With the popularity of cloud computing and micro service architectures, various service ecosystems including services, venders, and service-based processes continuously emerge on Internet or in an enterprise. Semantics of services from different venders may be described by distributed domain ontologies. Distributed knowledge brings difficulty to competition and cooperation among services, and hampers the evolution of a service ecosystem. In this paper, we propose a distributed knowledge based evolution model (DKEM) to promote competition and cooperation among services from different venders. DKEM considers stability as key factor in competition, and a stability evaluation model is designed to compute stability of services, venders, and service-based processes according to service invocation histories. Based on the evaluation model, two evolution patterns are given, and they can automatically explore new and more stable cooperation among services by means of runtime self-adaption mechanism. A prototype system for DKEM is implemented and a series of experiments show that DKEM is effective for competition and cooperation among services with distributed knowledge, and, evolved processes have higher stability and response efficiency.

Common software weaknesses, such as improper input validation, integer overflow, can harm system security directly or indirectly, causing adverse effects such as denial-of-service, execution of unauthorized code. Common Weakness Enumeration (CWE) maintains a standard list and classification of common software weakness. Although CWE contains rich information about software weaknesses, including textual descriptions, common sequences and relations between software weaknesses, the current data representation, i.e., hyperlined documents, does not support advanced reasoning tasks on software weaknesses, such as prediction of missing relations and common consequences of CWEs. Such reasoning tasks become critical to managing and analyzing large numbers of common software weaknesses and their relations. In this paper, we propose to represent common software weaknesses and their relations as a knowledge graph, and develop a translation-based, description-embodied knowledge representation learning method to embed both software weaknesses and their relations in the knowledge graph into a semantic vector space. The vector representations (i.e., embeddings) of software weaknesses and their relations can be exploited for knowledge acquisition and inference. We conduct extensive experiments to evaluate the performance of software weakness and relation embeddings in three reasoning tasks, including CWE link prediction, CWE triple classification, and common consequence prediction. Our knowledge graph embedding approach outperforms other description- and/or structure-based representation learning methods.

The study of paraconsistent reasoning with ontologies is especially important for the Semantic Web since knowledge is not always perfect within it. However, classical OWL reasoners cannot support reasoning with inconsistent ontologies. In this article, the authors present a plugin-based framework called prose to provide rich paraconsistent reasoning services for OWL ontologies, whose architecture contains the three following parts: a classical OWL reasoner, a multi-valued transformer, and an OWL API connecting with them. Within the proposed framework prose, they implement different multi-valued paraconsistent reasoning in the OWL. Moreover, they select three popular classical OWL reasoners and two typical kinds of reasoning services for users. As the authors excepted, prose does exactly enable current classical OWL reasoners to tolerate inconsistency in a simple and convenient way. Finally, they evaluate the three reasoners in a united framework (prose) and, as a result, those results can amend the analysis of the three reasoners on inconsistent ontologies.

The integration of cloud and Internet of Things (IoT), named CloudIoT, has been considered as an enabler for many different applications. However, the suspicion about the security issue is one main concern that some organizations hesitate to adopt such technologies while some just ignore the security issue while integrating the CloudIoT into their business. Therefore, given the numerous choices of cloud-resource providers and IoT devices, how to evaluate their security level becomes an important issue to promote the adoption of CloudIoT as well as reduce the business security risks. To solve this problem, considering the importance of the business data in CloudIoT, we develop an end-to-end security assessment framework based on software defined network (SDN) to evaluate the security level for the given CloudIoT offering. Specially, in order to simplify the network controls and focus on the analysis about the data flow through CloudIoT, we develop a three-layer framework by integrating SDN and CloudIoT, which consists of 23 different indicators to describe its security features. Then, the interviews from industry and academic are carried out to understand the importance of these features for the overall security. Furthermore, given the relevant evidences from the CloudIoT offering, the Google Brillo and Microsoft Azure IoT Suite, our framework can effectively evaluate the security level which can help the consumers for their CloudIoT selection.

Transboundary has become not only a trend but also an important requirement and feature of modern service industry development. In the Internet era, transboundary service not only changed people's daily life, but also had a profound impact on the business model of traditional industry. China has proposed the “Internet+” development plan, hoping to use the Internet for the transformation of traditional industries, including retail, manufacturing, catering, agriculture, finance, and health care. However, the transboundary impact of Internet on the operating model of traditional industry is decided by a combination of a variety of factors. Currently, the appropriate assessment model and analysis methods are absent in this field. Based on this, the impact evaluation method of Internet (service bridge) is proposed from the prospect of supply and demand matching, which includes three main parts: the capability model of supply side, the characteristic model of demand side, and the service bridge model. Based on the proposed method, the corresponding computational experiment system is built to evaluate the impact of Internet mode in different industries. Finally, this paper verifies the method with actual cases, and compares the transboundary impact of Internet in different daily consumption industries (online to offline in beauty service and take-out food service industry). The results showed that the “service bridge” method can introduce a new idea for the transboundary impact evaluation of Internet, which can provide some decision support to the reconstruction of demand value chain in some traditional industries.

In order to tackle the incomplete and inaccurate of the samples in most tracking-by-detection algorithms, this paper presents an object tracking algorithm, termed as multi-scale spatio-temporal context (MSTC) learning tracking. MSTC collaboratively explores three different types of spatio-temporal contexts, named the long-term historical targets, the medium-term stable scene (i.e., a short continuous and stable video sequence), and the short-term overall samples to improve the tracking efficiency and reduce the drift phenomenon. Different from conventional multi-timescale tracking paradigm that chooses samples in a fixed manner, MSTC formulates a low-dimensional representation named fast perceptual hash algorithm to update long-term historical targets and the medium-term stable scene dynamically with image similarity. MSTC also differs from most tracking-by-detection algorithms that label samples as positive or negative, it investigates a fusion salient sample detection to fuse weights of the samples not only by the distance information, but also by the visual spatial attention, such as color, intensity, and texture. Numerous experimental evaluations with most state-of-the-art algorithms on the standard 50 video benchmark demonstrate the superiority of the proposed algorithm.

In this article, we present a sequential approach for articulated motion estimation from a 2D skeleton sequence. This is a challenging task due to the complexity of human movements and the inherent depth ambiguities. The proposed approach models the human movement on a kinematic manifold with the tangent bundle, which is a natural geometrical representation of articulated motion. Combined with a second-order stochastic dynamic model based on the Markov hypothesis, we generalize the Extended Rauch Tung Striebel smoother to a Riemannian manifold to simulate the process of human movement. The human motor system might violate the Markov hypothesis when the human body is subject to external forces, and therefore a refinement stage is introduced to correct the estimation error. Specifically, the current estimation is refined in a feasible solution region consisting of a set of local estimations. This region is called a simplex, in which each element can be represented by a convex hull of all ingredients. We have proved that the refinement problem can be converted into a convex optimization problem with the simplicial constraint. Since the proposed formulation conforms to the principles of kinematic and spatio-temporal continuity of articulated motion, the reconstruction ambiguity can be alleviated essentially. The performance of the proposed algorithm is conducted on multiple synthetic sequences from the CMU and the HDM05 MoCap databases. The results show that, without requiring any training data, the proposed approach achieves greater accuracy over state-of-the-art baselines. Furthermore, the proposed approach outperforms two baselines on real sequences from the Human3.6m MoCap database.

The objective of discovering network communities, an essential step in complex systems analysis, is two-fold: identification of functional modules and their semantics at the same time. However, most existing community-finding methods have focused on finding communities using network topologies, and the problem of extracting module semantics has not been well studied and node contents, which often contain semantic information of nodes and networks, have not been fully utilized. We considered the problem of identifying network communities and module semantics at the same time. We introduced a novel generative model with two closely correlated parts, one for communities and the other for semantics. We developed a co-learning strategy to jointly train the two parts of the model by combining a nested EM algorithm and belief propagation. By extracting the latent correlation between the two parts, our new method is not only robust for finding communities and semantics, but also able to provide more than one semantic explanation to a community. We evaluated the new method on artificial benchmarks and analyzed the semantic interpretability by a case study. We compared the new method with eight state-of-the-art methods on ten real-world networks, showing its superior performance over the existing methods.

The Man-In-The-Middle (MITM) attack is one of the most common attacks employed in the network hacking. MITM attackers can successfully invoke attacks such as denial of service (DoS) and port stealing, and lead to surprisingly harmful consequences for users in terms of both financial loss and security issues. The conventional defense approaches mainly consider how to detect and eliminate those attacks or how to prevent those attacks from being launched in the first place. This paper proposes a game-theoretic defense strategy from a different perspective, which aims at minimizing the loss that the whole system sustains given that the MITM attacks are inevitable. We model the interaction between the attacker and the defender as a Stackelberg security game and adopt the Strong Stackelberg Equilibrium (SSE) as the defender's strategy. Since the defender's strategy space is infinite in our model, we employ a novel method to reduce the searching space of computing the optimal defense strategy. Finally, we empirically evaluate our optimal defense strategy by comparing it with non-strategic defense strategies. The results indicate that our game-theoretic defense strategy significantly outperforms other non-strategic defense strategies in terms of decreasing the total losses against MITM attacks.

With the advent of Web services, service interoperability has always been an active research issue. In recent years, many approaches have been proposed. However, how to achieve fast composition and guarantee correct and executable composite service remains an open issue. For this problem, this paper presents a three-phase framework for accurate and efficient service interoperability. Since service collaborations should follow certain constraints for success invocation, the goal of the first phase is to automatically clarify constraints on Web services. And then the second phase utilizes constraints acquired in previous phase to check Web services' constraint compatibility for accurate collaborations. In order to reduce the time on exhaustive analysis of service matching, the concept of expanded parameters is proposed, thus the problem of semantic matching is transformed into set operation. Subsequently, the third phase achieves interoperability among Web services by constructing initial composite services on the basis of collaborations, optimizing initial compositions to generate minimal composition alternatives with no redundant Web services, and executing final composition services. Experimental results show that our framework can dramatically reduce the time spending on service matching and effectively generate minimal composition alternatives in a rather short time.

Pre-learnt subspace methods, e.g., 3DMMs, are significant exploration for the synthesis of 3D faces by assuming that faces are in a linear class. However, the human face is in a nonlinear manifold, and a new test are always not in the pre-learnt subspace accurately because of the disparity brought by ethnicity, age, gender, etc. In the paper, we propose a parametric T-spline morphable model (T-splineMM) for 3D face representation, which has great advantages of fitting data from an unknown source accurately. In the model, we describe a face by C^2 T-spline surface, and divide the face surface into several shape units (SUs), according to facial action coding system (FACS), on T-mesh instead of on the surface directly. A fitting algorithm is proposed to optimize coefficients of T-spline control point components along pre-learnt identity and expression subspaces, as well as to optimize the details in refinement progress. As any pre-learnt subspace is not complete to handle the variety and details of faces and expressions, it covers a limited span of morphing. SUs division and detail refinement make the model fitting the facial muscle deformation in a larger span of morphing subspace. We conduct experiments on face scan data, kinect data as well as the space-time data to test the performance of detail fitting, robustness to missing data and noise, and to demonstrate the effectiveness of our model. Convincing results are illustrated to demonstrate the effectiveness of our model compared with the popular methods.

The widely used TLS protocol is vulnerable to Man-in-the-Middle (MITM) attacks in public WiFis. Such attacks arise since most users are often unable to verify server certificates properly and even worse, the implement of client authentication is typically decoupled from TLS session establishment. These two authentication procedures could be bound cryptographically in the establishment of TLS sessions, and a TLS security-enhanced mechanism (TLSsem) is proposed to detect and defense MITM attacks in public WiFis. TLSsem deals with the TLS mutual authentication through a way that combines pre-binding with certificate validation. Servers take advantage of pre-binding to generate identity credentials for users as pre-shared keys. Afterwards, the mutual authentication between clients and servers is realized in certificate validation by using the identity credentials, and also the forged server certificate is detected on the server side by checking this certificate. In addition, to ensure the reliability of wireless communications in this malicious public WiFis, an TLS shared service based on random port hopping is implemented to reallocate the reliable ports for data transmission against the interception by MITM attackers. We implement a prototype and verify its effectiveness by a thorough set of experiments in the real network environment. Evaluation results show that our mechanism can significantly increase the security of TLS communication in public WiFis without introducing noticeable overhead.

Software vulnerabilities pose significant security risks to the host computing system. Faced with continuous disclosure of software vulnerabilities, system administrators must prioritize their efforts, triaging the most critical vulnerabilities to address first. Many vulnerability scoring systems have been proposed, but they all require expert knowledge to determine intricate vulnerability metrics. In this paper, we propose a deep learning approach to predict multi-class severity level of software vulnerability using only vulnerability description. Compared with intricate vulnerability metrics, vulnerability description is the "surface level" information about how a vulnerability works. To exploit vulnerability description for predicting vulnerability severity, discriminative features of vulnerability description have to be defined. This is a challenging task due to the diversity of software vulnerabilities and the richness of vulnerability descriptions. Instead of relying on manual feature engineering, our approach uses word embeddings and a one-layer shallow Convolutional Neural Network (CNN) to automatically capture discriminative word and sentence features of vulnerability descriptions for predicting vulnerability severity. We exploit large amounts of vulnerability data from the Common Vulnerabilities and Exposures (CVE) database to train and test our approach.

Android system has been the crucial platform for the mobile service ecosystem. As a typical open source project, the release of the android system is a challenging issue because many developers are working on the related projects and it will affect millions of mobile service running on the platform. Therefore, investigating the release process of Android system is important for the mobile service ecosystem. Particularly, in this paper, we will focus on the release features prediction issue of what features should be included in the new publishing version. The valid changes and release notes are transformed into low-dimensional vectors and then the automatic labelling methodology is developed to detect the features. Combing with the time series forecasting model, an approach to predict the published features in the new version is presenting. Based on the data collected from the Android Open Source Project (AOSP), the experiments show that: comparing with the state-of-the-art, our approach achieves 13.83% to 17.69% precision improvement in releasing feature predictions and we can effectively detect the spike features for further compatibility management.

The Man-in-the-Middle (MITM) attack has become widespread in networks nowadays. The MITM attack would cause serious information leakage and result in tremendous loss to users. Previous work applies game theory to analyze the MITM attack-defense problem and computes the optimal defense strategy to minimize the total loss. It assumes that all defenders are cooperative and the attacker know defenders' strategies beforehand. However, each individual defender is rational and may not have the incentive to cooperate. Furthermore, the attacker can hardly know defenders' strategies ahead of schedule in practice. To this end, we assume that all defenders are self-interested and model the MITM attack-defense scenario as a simultaneous-move game. Nash equilibrium is adopted as the solution concept which is proved to be always unique. Given the impracticability of computing Nash equilibrium directly, we propose practical adaptive algorithms for the defenders and the attacker to learn towards the unique Nash equilibrium through repeated interactions. Simulation results show that the algorithms are able to converge to Nash equilibrium strategy efficiently.

Fifth-generation (5G) control/user (C/U) plane split heterogeneous network may cause more serious handover problems than traditional networks, especially for the inter-macrocell handover. In addition, the mobility behavior of mobile node (MN) may also result in improper handover triggers. In this paper, an adaptive handover trigger strategy (AHTS) is proposed to predict the received signal strength indicator (RSSI), to guarantee that handover trigger is accurate and timely when the MN is moving. First, the fast dynamic time warping (FastDTW) algorithm as well as the density-based spatial clustering of applications with noise (DBSCAN) algorithm are used for clustering trajectories to discover the mobility patterns of MN. Then multiple hidden markov models (HMM) are trained based on the results of clustering. Finally, according to the HMM model which is selected based on the similarity between test trajectory and each cluster center, an adaptive RSSI predication (ARP) algorithm is implemented to predict the values of the RSSI. Simulation results show that our strategy outperforms the existing schemes in terms of the handover probability and the handover success probability.

It is critical and foremost to come up with the corresponding security requirements first which the following implementations are based on. However, previous security requirement elicitation work based on Common Criteria (CC) rarely addresses the detailed elicitation process of threats from specific functional requirements, which thus results in the widen gap between specific functional requirements and their corresponding threats. To this end, this paper proposes a framework for eliciting corresponding security requirements of specific functional requirements from the requirements specification. A formal model is built in the framework to assist requirement analysts in half-automatic collecting threats. To enhance the framework's automaticity and reusability, a security property base is constructed based on authoritative sources of security properties to support the framework. A practical information system is applied to verify the framework's practicability. Finally the framework's advantages and limitations are discussed thoroughly compared with previous approaches and useful insights are revealed.

Since security is recommended to be evaluated at the beginning of the software development process, specifying software security requirements is inevitable in developing Critical Information Security Systems.However, according to the ISO/IEC 15408 (known as Common Criteria), determining detailed software security requirements (SRs) is quite challenging, complex which needs lots of expert knowledge of security.In this paper, a data-driven Functionality Topic-Security Requirement (FT-SR) model is proposed to recommend software SRs based on the relationship between software functionality specification and SRs from software Security Target (ST) which have been written in accordance with ISO/IEC 15408.First, we extract descriptions of functionality and tag SRs all from software STs.Second, Latent Dirichlet Allocation (LDA) is adopted to build functionality topics for product functionality description.Third, a FT-SR model is developed based on the mapping between product functionality topics and SRs which have been tagged in ST documents.Finally, a recommendation strategy is proposed to recommend SRs based on the FT-SR model for software products.Our experiments are performed on ST documents of over 600 software products provided by Common Criteria.Experimental results show that the proposed approach can generate a set of recommended SRs reducing the difficulty of SRs recommending even for people lack knowledge of security.

In this paper, we present a federated path querying language (FPQ) as itself an extension of the nested regular path querying language with adding an axis operator to support the federation of RDF dataset and relational database. We have proven that FPQ has more expressive power than the nested regular path query language (not to mention regular path querying language). It enjoys the same computational complexity as the regular path query language and its additional expressivity can be exactly used to characterize the conjunction and federation of nested regular path queries. Moreover, we discuss the expressivity of various fragments of FPQ and implement FPQ. Finally, we present an application scenario related to the car-pooling services in real life.

Summary For service‐based process, there is a frequent failure when it runs under a loose‐coupled and uncertain environment, such as mobile Internet. The runtime self‐adaptation mechanism is urgently needed to automatically achieve the goal of a process. Therefore, this article presents a self‐adaptation framework to automatically monitor and self‐adapt to various failures. On the basis of the presented self‐adaptation exceptions framework, the service‐based process is automatically extended with these exception definitions. In particular, the local self‐adaptation exceptions monitor the failures over each web service instance, whereas the global self‐adaptation exceptions are designed to explore the process‐level failures. For each exception, the self‐adaptation strategy is dynamically generated according to the running environment. Finally, we implement the prototype system running on the Apache ODE. The experimental results show that our framework can automatically recover a process instance from various failures, and the proposed approach can identify more failure types than previous works and apparently improve the running success rate of process instances.

In multiagent systems the capability of learning is important for an agent to behave appropriately in face of unknown opponents and a dynamic environment. From the system designer's perspective, it is desirable if the agents can learn to coordinate towards socially optimal outcomes, while also avoiding being exploited by selfish opponents. To this end, we propose a novel gradient ascent based algorithm (SA-IGA) which augments the basic gradient-ascent algorithm by incorporating social awareness into the policy update process. We theoretically analyze the learning dynamics of SA-IGA using dynamical system theory, and SA-IGA is shown to have linear dynamics for a wide range of games including symmetric games. The learning dynamics of two representative games (the prisoner's dilemma game and coordination game) are analyzed in detail. Based on the idea of SA-IGA, we further propose a practical multiagent learning algorithm, called SA-PGA, based on the Q-learning update rule. Simulation results show that an SA-PGA agent can achieve higher social welfare than previous social-optimality oriented Conditional Joint Action Learner (CJAL) and also is robust against individually rational opponents by reaching Nash equilibrium solutions.

In the MANETs (Mobile Ad Hoc Networks), attack detection in Location Aided Routing (LAR) has become a challenging problem due to the dynamic network topology and diverse routing attacks. This paper proposes a universal framework to analyze and detect attack traces for various potential attack targets. In this framework, the attack targets are identified by matching key events, and a reverse search algorithm based on iteration is employed to obtain attack traces and attack conditions, which are guidance for generation of detection rules. Algebra for Wireless Mesh Networks (AWN) is extended to specify the core functionality of LAR, and detection rules are validated by NS-2 platform. The results demonstrate that the approach is correct and efficient.

Community detection has been one of the key research topics in the analysis of networked data, which is a powerful tool for understanding organizational structures of complex networks. One major challenge in community detection is to analyze community structures for streaming networks in real-time in which changes arrive sequentially and frequently. The existing incremental algorithms are often designed for edge-grained sequential changes, which are sensitive to the processing sequence of edges. However, there exist many real-world net-works that changes occur on node-grained, i.e., node with its connecting edges is added into network simultaneously and all edges arrive at the same time. In this paper, we propose a novel incremental community detection method based on modularity optimization for node-grained streaming networks. This method takes one vertex and its connecting edges as a processing unit, and equally treats edges involved by same node. Our algorithm is evaluated on a set of real-world networks, and is compared with several representative incremental and non-incremental algorithms. The experimental results show that our method is highly effective for discovering communities in an incremental way. In addition, our algorithm even got better results than Louvain method (the famous modularity optimization algorithm using global information) in some test networks, e.g., citation networks, which are more likely to be node-grained. This may further indicate the significance of the node-grained incremental algorithms.

In this paper, we analyze some basic features of SPARQL queries from practical world in a statistical way. In particular, we focus on three statistic features including the occurrence frequency of triple patterns, fragments, and well-designed patterns and four semantic features including monotonicity, non-monotonicity, weak monotonicity and satisfiability. All the features contribute to characterize SPARQL queries in different dimensions. We hope that this statistical analysis would provide some useful observations for researchers and engineers who are interested in what real-word SPARQL queries look like, so that they could develop some practical heuristics for processing SPARQL queries, as well as build SPARQL query processing engines and benchmarks. In addition, our research facilitates to reduce scope of the problems by avoiding some cases that may not occur in practice.

Ensuring the fairness and non-repudiation in the security exchange protocol of web service is critical. Model checking is often used for automatic verification for the security properties of protocol. However, the current model checker tools cannot support formalizing protocols with cryptographic primitives, specifying properties with linear temporal logic (LTL) and automatically generating resilient intruder model simultaneously and the application range of them is severely limited. To solve this problem, a model checker Fepchecker is proposed to verify the fairness and non-repudiation properties, which are critical features in security exchange protocols. Firstly, applied pi-calculus is extended to specify the protocols, and the LTL assertion is used for precisely describing fairness and non-repudiation. Secondly, an intruder model is applied to construct their behavior sequences automatically and the protocol sessions and message pattern are used to alleviate the states explosion problem. Thirdly, in our model checking algorithm, the fairness and non-repudiation properties are verified based on Labeled Transition System (LTS) semantics model and the MakeOneMove method is used to explore the state space on-the-fly in the verification process. Finally, Fepchecker is applied to verify six representative protocols and the results show that Fepchecker can effectively verify their fairness and non-repudiation properties.

By leveraging online social networks as an underlying infrastructure, Social Participatory Network (SPN) has been becoming a new paradigm of participatory sensing systems. However, a significant barrier to the widespread use of SPN applications is their vulnerability to various forms of malicious attacks. Such threats inhibit human participation and thus the viability of SPN systems in everyday use. To solve this problem, this paper proposes a trust evaluation framework for participants to encourage wider human participation in SPN. The proposal is based on the Tianjin University's own existing SPN system, named CRCS (ClassRoom Cloud System), which enables participants to use the cloud resources for online lessons or library study. It derives the trust value of participants by using entropy-weight method and data mining algorithms to deal with the behaviors data of participants. Our proposed solution can detect malicious participants easily, and more importantly, it outperforms other work for its low cost and simple deployment. For now, though our solution is based on a specified SPN system, we are confident that this solution is highly applicable to most other SPN systems.

In this paper, we propose an architecture security evaluation method to identify potential risks of architecture. We model security features of web applications from two different but complementary points of view using AADL, build an AADL security model which contribute to detect risks of architecture. With the help of a tool, we can automatically convert the AADL security model to an architecture security model. Then, an integration process applies analytic hierarchy process (AHP) and fuzzy evaluation analysis to the architecture security model. In the end, we can get security conclusions of the architecture and improve security measures based on security conclusions. The experiment demonstrates that the method not only improves efficiency of the evaluation, but also makes security evaluation process more objective and accurate.

Semantic annotation plays an important role for semantic-aware web service discovery, recommendation and composition. In recent years, many approaches and tools have emerged to assist in semantic annotation creation and analysis. However, the Quality of Semantic Annotation (QoSA) is largely overlooked despite of its significant impact on the effectiveness of semantic-aware solutions. Moreover, improving the QoSA is time-consuming and requires significant domain knowledge. Therefore, how to verify and improve the QoSA has become a critical issue for semantic web services. In order to facilitate this process, this paper presents a novel lifecycle framework aiming at QoSA assessment and optimization. The QoSA is formally defined as the success rate of web service invocations, associated with a verification framework. Based on a local instance repository constructed from the execution information of the invocations, a two-layer optimization method including a local-feedback strategy and a global-feedback one is proposed to improve the QoSA. Experiments on real-world web services show that our framework can gain 65.95%~148.16% improvement in QoSA, compared with the original annotation without optimization.

Loosely coupled and cross-platform features make Web services accessible and increasingly popular on the Internet. However, efficient service discovery and automated service composition are still challenges under the conventional practice where services are organized into categories. In this paper, we propose a graph-based method to organize Web services into a service ecosystem interlaced with service relationships at the semantic level. First, Web services are modelled as a set of interfaces, whose input and output parameters are annotated with well-defined ontologies. Secondly, semantic associations and interactions between Web services are mined, and services are constructed into a Web services network (SN), a variant of bipartite graph, by projecting the functional aspects of concrete Web services onto the abstract service layer. Thirdly, from the complex network perspective, the services relations are investigated and the structure of SN is analysed. To demonstrate the basic topological properties of SN, an empirical study is conducted on two data sets for comparative purposes, 10 000+ Web services collected from the Internet and 1231 Web services provided by Titan system of Zhejiang University. The experimental results reveal that SNs, which are built by different data sets on the semantic level, exhibit the same features such as small-world and scale-free. In addition, our results yield valuable insight for developing service discovery and automated composition algorithms, and characterizing the evolution of the entire Web service ecosystem.

Facial expressions can be mainly conveyed by only a few discriminative facial regions of interest. In this paper, we study the discriminative regions for facial expression recognition from video sequences. The goal of our method is to explore and make use of the discriminative regions for different facial expressions. For this purpose, we propose a Hidden Markov Model (HMM) Decision Forest (HMMDF). In this framework, each tree node is a discriminative classifier, which is constructed by combining weighted HMMs. Motivated by a psychological theory of "elimination by aspects", several HMMs on each node are modeled respectively for facial regions, which have discriminative capabilities for classification. The weights for these HMMs can be further adjusted according to the contributions of facial regions. Extensive experiments validate the effectiveness of discriminative regions on facial expression, and the experimental results show that the proposed HMMDF framework yields dramatic improvements in facial expression recognition compared to existing methods.

This paper presents a formal modeling method of security protocols based on the extended UML framework. In order to simplify the process and reduce the difficulty of security protocol modelling, extending mechanisms for the class diagram and sequence diagram of UML are presented, which provide an engineering specification for the security protocol formalizing. Therefore, for verifying the confidentiality and correspondence of security protocols by Prover if, a transformation from extended UML model to Prover if Spi calculus model is realized with matching rules and knowledge reasoning, and then the verifying results are analyzed through a regular expression. Finally, the handshake, NS public key and buyer-seller watermarking protocols are verified, the attack traces of unsatisfied security properties are exported, that show the validity and applicability of the approach provided by this paper.

This paper proposes a Software Risk Assessment Method based on Object-Oriented Petri Net (OOPN-SRAM), in which risk assessment procedure is divided into four steps, expressed as four corresponding objects, including asset recognition, weakness analysis, consequence property confirmation and risk calculation. Each object is modeled with Petri net. Specialists recognize software assets by the 1-9 scales method of Analytic Hierarchy Process (AHP). The weaknesses in a system are found by the vulnerability scanner. The damage degree and the exploitation likelihood of a weakness are evaluated by such authorities as Common Weakness Enumeration (CWE). The consequence properties are confirmed by specialists according to the software requirements. Finally, in the risk calculation, risk degree and overall risk value are calculated by using exponential method and weighted average method respectively. Furthermore, we illustrate the application of our OOPN-SRAM method with realistic examples including web-banking and forum, and make a comparison with traditional methods. The results show that OOPN-SRAM not only increases the efficiency of the evaluation process, but also makes the evaluation result more objective and accurate.

Service relations facilitate the automation of service reuse. Most of studies on the service relations focus on the inputs and outputs. However, different Web Services tend to utilize the same parameters without formally specifying their constraints. Due to this, the semantics, introduced by semantic annotation, is still not rich enough for accurate descriptions, thus generating a large number of inappropriate service relations. To address this, we propose an approach for augmenting semantics of Web Services based on constraints, which can be regarded as a complement to semantic annotation. The semantics is augmented via a hybrid analysis of heterogeneous constraints, including the server constraint and object constraint.

A major challenge in physiology-based emotion recognition is to establish an effective emotion recognizer for multi-users in the user-independent scenario. The recognition result is not satisfied because it ignores the difference in individual response pattern, which can be attributed to IRS (Individual Response Specificity) and SRS(Stimuli Response Specificity) in psychophysiology. To improve the performance of emotion recognition, this paper proposes a Group-Based IRS model by adaptively matching a suitable recognizer in accordance with user's IRS level. Specifically, the users are put into distinct groups by using cluster analysis techniques, where users within the same group have similar IRS level than other groups. Then physiological data of users from each group is utilized to build the corresponding emotion recognizers. After categorizing a new user into one group according to his IRS level, the new user's emotion state is predicted by the corresponding emotion recognizer. To validate our model, the affective physiological data was collected from 11 subjects in four induced emotions(neutral, sadness, fear and pleasure), three-channel bio-sensors were used to measure users electrocardiogram (ECG), galvanic skin response (GSR) and photo plethysmography (PPG). The results show that the recognition precision in Group-based IRS model is higher than general model.

This paper mainly studies facial expression recognition with the components by face parsing (FP). Considering the disadvantage that different parts of face contain different amount of information for facial expression and the weighted function are not the same for different faces, an idea is proposed to recognize facial expression using components which are active in expression disclosure. The face parsing detectors are trained via deep belief network and tuned by logistic regression. The detectors first detect face, and then detect nose, eyes and mouth hierarchically. A deep architecture pretrained with stacked autoencoder is applied to facial expression recognition with the concentrated features of detected components. The parsing components remove the redundant information in expression recognition, and images don't need to be aligned or any other artificial treatment. Experimental results on the Japanese Female Facial Expression database and extended Cohn-Kanade dataset outperform other methods and show the effectiveness and robustness of this algorithm.

Software security evaluation is considered as a significant and indispensible activity in all phases of software development lifecycle, and there are also many factors that should be taken into account such as the environment, risks, and development documents. Despite the achievements of the past several decades, there is still a lack of methodology in evaluating software security systematically. In this paper, we propose a comprehensive model for evaluating the software security from three different but complementary points of view: technology, management and engineering. The technological dimension is 7 security levels based on Evaluation Assurance Levels (EALs) from ISO/IEC15408, the management dimension mainly concerns the management of software infrastructures, development documents and risks, and the engineering dimension focuses on 5 stages of software development lifecycle. Experts evaluate software security through the evidence items which are collected from these three dimensions and provide their assessments. Relying on Analytic Hierarchy Process (AHP) and Dempster-Shafer Evidence Theory, assessments obtained from the experts can be combined and merged to get a score which presents the security degree of software. A case study illustrates how the evaluators may use the proposed approach to evaluate security of their system.

In this paper, a method about how to identify insecure behaviors of browser extensions is proposed. Typically, the identification of insecure extension behaviors is based on knowledge which is got by investigating known malicious or vulnerable extensions. We present an automatic technique that can ease the laborious manual investigating process. Our technique mines the difference between the behavior graphs of insecure and secure extensions based on graph mining algorithm. The difference between them is the specification of insecure extension behaviors which can be further analyzed manually or automatically to help people make a better decision about whether an extension is secure or not. We developed a prototype and the experimental results show that this kind of technique can effectively extract insecure extension behaviors.

This paper proposes an Android malware detection approach based on attack tree. Attack tree model is extended to provide a novel way to organize and exploit behavior rules. Connections between attack goals and application capability are represented by an attack tree structure and behavior rules are assigned to every attack path in the attack tree. In this way, fine-grained and comprehensive static capability estimation and dynamic behavior detection can be achieved. This approach employs a hybrid static-dynamic analysis method. Static analysis tags attack tree nodes based on application capability. It filters the obviously benign applications and highlights the potential attacks in suspicious ones. Dynamic analysis selects rules corresponding to the capability and conducts detection according to runtime behaviors. In dynamic analysis, events are simulated to trigger behaviors based on application components, and hence it achieves high code coverage. Finally, in this way, we implement an automatic malware detection prototype system called AM Detector. The experiment result shows that the true positive rate is 88.14% and the false positive rate is as low as 1.80%.

ABSTRACT Design‐level vulnerabilities are a major source of security problems in software programs. For the purpose of improving the trustworthiness of software designs, this paper presents a unified threat model for representing, analyzing, and evaluating software threats at various design stages. Unified threat models represent software threats via tree structures with AND/OR logical relationships and evaluates software threats in a cost‐effective way based on attack paths. Mitigation measures for software threats are designed and prioritized based on the evaluation results, which make it possible to design high‐quality software security programs that resist identified software threats. A case study for an online banking system is given to systematically demonstrate the application of unified threat models in software threat analysis and evaluation. The results from the case study demonstrate that the unified threat model is superior to traditional threat trees in accurately evaluating results, designing mitigation measures, and guiding software security testing. Copyright © 2012 John Wiley & Sons, Ltd.

This paper presents a quantitative architecture security evaluation method to identify potential risks of an architecture. The method is based on security of the architecture components. In this method, components of the architecture are classified and their security measures are identified according to component function and architecture level. Then, an integration process applies analytic hierarchy process (AHP) and fuzzy evaluation analysis to determine quantitative and qualitative factors in evaluating the security of components. These factors are used to obtain security conclusions of the architecture. The experiment shows that the method not only improves efficiency of the evaluation, but also makes security evaluation process more objective and accurate.

Large scale ontology applications require efficient reasoning services, of which ontology classification is the fundamental reasoning task. The special EL reasoners are efficient, but they can not classify ontologies with axioms outside the OWL 2 EL profile. The general-purpose OWL 2 reasoners for expressive Description Logics are less efficient when classifying the OWL 2 EL ontologies. In this work, we propose a novel technique that combines an OWL 2 reasoner with an EL reasoner for classification of ontologies expressed in DL SROIQ. We develop an efficient task decomposition algorithm for identifying the minimal non-EL module that is assigned to the OWL 2 reasoner, and the bulk of the workload is assigned to the EL reasoner. Furthermore, this paper reports on the implementation of our approach in the ComR system which integrates the two types of reasoners in a black-box manner. The experimental results show that our method leads to a reasonable task assignment and can offer a substantial speed up (over 50%) in ontology classification.

We present a new approach to characterizing the semantics for the integration of rules and first-order logic in general, and description logics in particular, based on a circumscription characterization of answer set programming, introduced earlier by Lin and Zhou. We show that both Rosati's semantics based on NM-models and Lukasiewicz's answer set semantics can be characterized by circumscription, and the difference between the two can be seen as a matter of circumscription policies. This approach leads to a number of new insights. First, we rebut a criticism on Lukasiewicz's semantics for its inability to reason for negative consequences. Second, our approach leads to a spectrum of possible semantics based on different circumscription policies, and shows a clear picture of how they are related. Finally, we show that the idea of this paper can be applied to first-order general stable models.

Social annotation provides a convenient way to annotate shared content by allowing users to use any tag or keyword. While free folksonomy is widely used in social software implementations and especially in web services, it will play an important role in the semantic web services. However, such tags cannot offer the expressivity of ontologies, and the respective tags often lack context-independent and explicit semantic. In this paper, we describe a model to extract domain concept from social tags. The model mainly includes three modules: a) Detecting the noun terminology through mutual information, b) Applying semantic dictionary to disambiguate between tags, c) Filtering the domain concept via domain relevance and consensus. Finally, experimental results on real world data sets show that the model can effectively learn the domain concept from social tags, and the concept also has a high degree of generality and applicability.

In this work, we describe a new technique for detecting image composites by enforcing two-view geometrical constrains: H and F constraints on image pairs, where H denotes the planar homography matrix and F the fundamental matrix. Our approach detects fake regions efficiently on pictures taken at the same scene but with different camera configurations. Performance of this approach is demonstrated on real image pairs with visually plausible composites.