An accurate perception of traffic environments is essential for safe autonomous driving. LiDAR point cloud semantic segmentation, an essential capability, is a key technology for accurately perceiving traffic elements. Feature extractions for LiDAR semantic segmentation present difficulties due to several factors. To cite a typical example, the vast-scale variations in objects in complex traffic scenes make it difficult to simultaneously capture the fine-grained details of small objects and the contextual information of large objects. To address these types of issues, this paper proposes a Multiscale Agent-gated Sensor Fusion (MASF) method, which projects 3D point clouds into 2D feature maps, multiscale feature extraction, and adaptive cross-modal fusion mechanisms. First, a 3D point cloud is projected into 2D feature maps, and its features are extracted alongside camera images in a dual-stream encoder. At each encoder stage, the camera features are fused into the LiDAR stream. To refine the fused multimodal features, this paper develops a Multiscale Gated Bottleneck Convolution (MS-GBC) mechanism to adaptively select features across different spatial scales. Second, at the encoder’s bottleneck, the fusion process is handled by the Low-Rank Agent Attention (LRAA) mechanism, which introduces a compact set of agent tokens to capture global dependencies across modalities. Third, a decoder progressively upsamples the context-rich feature map from the encoder, reintegrating fine-grained details through skip connections. Then, a segmentation head generates the final segmentation results. Finally, experiments on two large-scale real-world datasets show that the performance of MASF is superior to that of many baseline methods. For example, it achieves mean intersection-over-union (mIoU) improvements of 2.6% and 2.0% over the best LiDAR-only methods on the two datasets.

Metal component manufacturing requires stringent surface quality standards to prevent structural failures in critical applications. Metal surface defect detection remains challenging in resource-constrained industrial edge environments. Highly textured, non-stationary backgrounds easily obscure tiny defects with weak visual saliency. Furthermore, such defects exhibit pronounced anisotropic geometry. Existing models struggle to achieve global semantic understanding, accurate geometric alignment, and real-time inference under limited computational budgets. These limitations lead to frequent detection failures. EDGL-Net is proposed as an adaptive multi-scale detection architecture for edge deployment. EDGL-Net integrates global context modeling and anisotropic geometric feature extraction. It incorporates a parameter-sharing multi-scale prediction mechanism to enhance robustness for small and elongated defects. Experiments on the NEU-DET and GC10-DET datasets show that the proposed method achieves a favorable balance between detection accuracy and computational efficiency. EDGL-Net improves [email protected] by 2.7 points and Precision by 6.0 points over the baseline on NEU-DET. It consumes 64% of the computational resources required by mainstream models.

The rapid expansion ofMobile Edge Computing(MEC) enables latency-sensitive applications by extending computation to the network periphery. Yet its distributed and dynamic nature poses significant challenges for secure key management and consistent trust governance. ConventionalPublic Key Infrastructure(PKI) schemes suffer from heavy cross-domain overhead and fragile certificate maintenance, revealing a fundamental mismatch between centralized trust management and decentralized operation. To overcome these limitations, we present PRISM (Policy-Regulated Identity-Based Secure Messaging), a decentralized security framework designed for dynamic MEC ecosystems. PRISM integratesIdentity-Based Encryption(IBE) andAttribute-Based Access Control(ABAC) to realize unified authentication and fine-grained authorization. In parallel, blockchain-based smart contracts enforce network-wide policies and ensure global integrity. A hierarchical design supports scalable coordination and adaptive key lifecycle management, while a two-tier on-chain governance layer maintains transparency and state consistency. Formal analysis and experiments show that PRISM effectively constrains on-chain storage overhead, achieves a 100% failure detection rate when the reporting window exceeds six rounds, and maintains a stable topology (Adjusted Rand Index, ARI > 0.8) under intensive churn. These results demonstrate that PRISM effectively reconciles decentralized control with secure coordination, offering a practical foundation for consistent trust governance in large-scale MEC environments.

Memory disaggregation has become a research trend in data centers. Existing studies fall into two paths: Network-based logical memory disaggregation (LMD) and Compute Express Link (CXL)-based physical memory disaggregation (PMD). However, LMD suffers from network overhead, while PMD incurs expensive hardware costs and lacks flexibility. This paper advocates for building LMD systems on shared CXL memory, taking advantage of its low latency and cache-coherent memory access. However, shared CXL memory has severe scalability issues due to its strict coherence model.This paper introduces Swarm, an efficient LMP system built on shared CXL memory. Swarm divides shared CXL memory into small cacheable memory and large non-cacheable memory. Hardware only needs to maintain coherence for cacheable memory, while software handles coherence for non-cacheable memory, thereby enabling all CXL memory to be shared. Then, Swarm implements cross-node RPC and dynamic global memory allocation on shared CXL memory to improve performance and memory utilization. Swarm also proposes distributed computing offloading to fully leverage compute power on memory nodes for acceleration. Our evaluation shows that Swarm not only improves the throughput (e.g., by 3.6× and 1.8×) compared with representative network-based LMD system, AIFM and CXL-based PMD system when computing offloading is enabled but also achieves an advantage in TCO.

The rapid development of the Internet of Vehicles (IoV) has accelerated technological progress, but several critical security challenges remain, especially in the context of vehicle trust management. Two representative issues are malicious nodes and unreliable information transmission. To address these problems, we propose BiTrustChain, a dual-layer blockchain framework designed to enhance security and trust management in IoV environments. First, it consists of two innovative data chains: a Behavior Data Chain (BDC) and a Reputation Evaluation Chain (REC). The BDC records vehicle interaction data, whereas the REC stores and updates the trust values in real time. Second, within this framework, we develop a Multifactor Bayesian Reputation (MFBR) model that enables quantitative evaluation of node trustworthiness. It integrates a time-decay function and a penalty mechanism to regulate reputation evolution. The trust values decrease after malicious behaviors and recover through continuous normal interactions. In addition, we propose a dynamic local whitelist for indirect reputation evaluation. It filters out untrustworthy nodes and ensures that only reliable nodes remain. The filtered indirect trust is then combined with direct trust to produce a comprehensive reputation score. Third, we design a new set of event-driven smart contracts to synchronize the BDC and REC in real time and ensure secure and efficient data exchange. Finally, we performed experiments on the evaluation platform SUMO/NS-3, and the results show that our method identifies malicious nodes with higher accuracy. In particular, the framework achieves 1.5× higher throughput and reduces latency by 40% compared to the baseline single-chain system. The framework also enhances interaction data integrity and improves robustness against adversarial reputation manipulation.

3D object detection plays a pivotal role in facilitating comprehensive scene understanding in autonomous driving systems. One of its key challenges is to achieve accurate perception in complex environments. Compared with LiDAR systems and stereo-vision approaches, monocular camera-based solutions are more cost-effective and easier to deploy. However, the absence of depth in monocular images hinders the accurate localization of 3D bounding boxes when only monocular images are used. This work proposes MonoLS, a monocular 3D object detection framework that incorporates lightweight multi-scale feature fusion and spatially-aware attention. It aims to address the challenge of missing depth information while achieving precise object localization. First, lightweight multi-scale feature fusion combines deep and shallow features. This design allows for effective multi-scale feature extraction without compromising real-time detection capabilities. Second, spatially-aware attention employs a dual-branch structure, with the spatial branch using a triplet attention to capture spatial details, and the context branch aggregating global context information through global attention. These two branches are subsequently fused to produce enhanced feature representations that preserve spatial distribution and semantic richness. Finally, experiments on the KITTI dataset demonstrate that our method outperforms the baseline, achieving a real-time inference speed of up to 67 FPS.

Efficient discovery and composition of services from large-scale cloud API repositories are fundamental to Mashup development and the rapid evolution of service-oriented software. However, extreme interaction sparsity severely limits existing recommendation methods, hindering their ability to infer unobserved relationships and capture complex service dependencies. Most prior approaches rely on deterministic point estimation in latent spaces, which leads to several critical issues under sparse supervision: difficulty in modeling preference uncertainty, high susceptibility to overfitting, and static and passive incorporation of multi-source auxiliary information. To address these challenges, we propose MS$^{2}$-Diff, a conditional diffusion-based generative framework for service recommendation that reformulates recommendation as a conditional probabilistic denoising process. MS$^{2}$-Diff has a three-fold main idea: 1) it characterizes two levels and four types of priors that drive Mashup development, including structural homophily and collaborative reachability at the structural level, as well as functional alignment and intent alignment at the semantic level; 2) it introduces a guidance-driven active retrieval mechanism via inverted attention within the denoising process, which actively retrieves and reconstructs latent relational patterns from noisy representations under the control of multi-view priors; and 3) it adopts a step-aware adaptive fusion strategy that dynamically coordinates structural and semantic priors across different denoising stages, enabling complementary and collaborative effects throughout the generation process. Extensive experiments on two real-world benchmark datasets demonstrate that MS$^{2}$-Diff achieves competitive and overall superior performance across multiple evaluation metrics, while improving robustness in sparse and cold-start scenarios.

In recent years, edge networks have been deployed broadly at large scale, hosting a wide variety of services. Among these, microservices have emerged as one of the predominant service paradigms. Typically, microservices run on edge servers with varying configurations, while new microservice instances are usually online generated and join in edge networks due to the dynamic attributes of requests and networks. In those cases, an effective microservice online deployment solution are expected to be vital to system performance. So it becomes a critical issue to design online deployment solutions for microservices in edge. Existing research has always focused on offline deployment of microservices. However, edge networks are characterized by dynamics, real time, and concurrency, and when the environments or requests change, traditional offline deployment solutions usually cannot handle the deployment task in such cases. To address these issues, we carry out a comprehensive investigation on those potential influencing factors in edge, fully covering deployment cost, load balance, packet loss, and network delay. We further develop an innovative holistic online deployment solution that encompasses a system model, constraint analysis, multiobjective optimization, and a deep reinforcement learning algorithm. We conducted extensive experiments and evaluated our solution over a set of metrics using a real-world microservice prototype system. The results show that our online deployment solution produces superior performance, for example, reducing deployment cost by an average of 70.14% compared to all baselines. We also evaluated our solution under varying volumes of requests and gave analysis for performance stability and parameter sensitivity. We have released the code on GitHub.

In the field of Web knowledge base mining and Web services, the recognition service for named entity faces many challenges such as context complexity, semantic subtlety, and fuzzy entity boundaries, all of which require highly accurate and robust recognition service. The current services usually fail to reach those conditions. To address this issue, this paper proposes a recognition service for named entities for large Web knowledge bases, and the core contrition is the developed dual multilayer feature learning (D-MLFL) service, which combines projected gradient descent (PGD), adversarial learning, and a fused attention mechanism. Our service successfully addresses the challenges of complex context and subtle semantics faced by named entity recognition tasks. Our service integrates deep language models, recurrent neural networks, and conditional random fields, and clearly outperforms existing approaches in many sub-tasks, including in feature extraction, sequence modeling, and label decoding, especially in dealing with complex and diverse entity types and contextual relationships. We performed sufficient experiments, and the results show that our service significantly enhances the robustness against noise and abnormal Web data. The ability to extract entity features is improved, resulting in higher accuracy in identifying entities with fuzzy boundaries and complex semantics.

In representation learning for Web temporal knowledge bases, each node in Web knowledge bases carries a specific contextual meaning. Existing services often neglect the implicit semantic Web knowledge behind entities and relations, thus failing to effectively capture the knowledge representation of temporal Web knowledge bases. To address this issue, this paper develops an autocompletion service for temporal Web knowledge bases, which is based on multisource semantic feature learning and feature fusion. We construct a semantic model oriented toward external semantic Web repositories to supplement entity-relation descriptions, and our service leverages the pretrained language model BERT, effectively learning semantic knowledge features. Additionally, our service captures the textual features of quadruples using a recurrent neural network, constructs a historical sparse timestamp matrix, and generates a mask tensor, successfully obtaining the weights of potentially correct entities, and thereby capturing the historical features of quadruples. Furthermore, our service integrates complementary features from different modules through an attention mechanism. Experimental validation shows that our service outperforms existing approaches in terms of four evaluation metrics: mean reciprocal rank (MRR), Hits@1, Hits@3, and Hits@10. The results also exhibit that it improves the accuracy and performance for autocompletion service for temporal Web knowledge bases.

Snapshot technology is crucial for data protection and system recovery in virtualized environments, particularly with the growing need for continuous snapshots to maintain the integrity of long-running web-based and distributed applications. However, traditional snapshot methods often suffer from performance bottlenecks, and inefficient storage usage. These challenges are closely tied to the way memory pages are accessed during VM execution, where memory access patterns show significant disparities between frequently accessed "hot" pages and less-used "cold" pages. In this paper, we introduce HeatSnap, a continuous snapshot system designed to address these issues by leveraging the uneven access frequencies of memory pages. HeatSnap distinguishes between intensive hot pages and dirty pages, applying specialized snapshotting and storage strategies to optimize the handling of both hot and cold memory regions. This approach aims to optimize snapshot efficiency, minimize performance impact on the VM, and decrease storage costs. Our implementation of HeatSnap on QEMU/KVM demonstrates significant improvements in VM performance loss, snapshot duration, and storage efficiency compared to existing methods, as evidenced by evaluations on common web and cloud-based workloads.

In the field of the industrial Internet, monitoring data from industrial equipment exhibit characteristics of high concurrency, high throughput, and high-frequency time series. Anomaly detection can accurately analyze the health status of equipment and enhance the monitoring capabilities of industrial Internet devices. To address the complex relationships between multi-attribute data and the need for anomaly detection, this paper presents an unsupervised multi-attribute industrial anomaly detection approach called the skip-range transformer (SRT). This approach learns anomaly features through parallel segmentation and skip-range attention to guide anomaly detection. First, each data point in the time series is transformed into a waveform graph, represented as a data graph representation (DGR), to capture key features such as temporal trends, periodicity, and abnormal points. By modeling multi-attribute time series data in parallel through the use of data graphs, the visual relationships, structures, and patterns among multi-attribute data are obtained. Second, our proposed approach jointly takes advantage of skip attention and range attention mechanisms to learn features from time series. Skip attention allows the model to capture dependencies by sampling at specified intervals, whereas range attention focuses on dividing the time series data within a specified time span, enabling the model to learn intricate features better. Third, the graph and data features are concatenated to form new features based on the new data generated from the self-attention mechanism, and then, anomalies are detected by evaluating the reconstruction error between the ground-truth time series data and the generated data. Finally, the experimental results demonstrate that the proposed approach outperforms the baseline methods, highlighting its ability to detect anomalies in industrial time series.

Cloud API recommender system has emerged as a promising solution to address the overload problem caused by the overwhelming growth of cloud APIs, aiming to improve software development efficiency. However, the incremental development nature of the service-oriented software necessitates an adaptive complementary cloud API recommender system, while existing systems purely focus on generating single-function, high-quality, and personalized recommendations based on retrieval content, quality of service, or user preferences, and overlook the practical scenarios that recommend high-order complementary cloud APIs for developers. This paper addresses this gap by providing a logical reasoning network for high-order complementary cloud API recommendation (LRN4HCAR). We first construct three relation graphs to characterize cloud API co-invocation, function co-occurrence, and substitute relations, capturing the strong complementarity, weak complementarity, and non-substitute relations among cloud APIs. Next, we develop a high-order complementary logical reasoning network with three sub-networks: strong complementary logical reasoning, weak complementary logical reasoning, and non-substitute logical reasoning. This network enables the recommendation of cloud APIs with high-order complementary relationships while excluding substitutes. Utilizing LRN4HCAR, we evaluate classic and SOTA baselines across various complementary recommendation scenarios on two real-world datasets. Experimental results demonstrate that LRN4HCAR outperforms others in all experimental settings for complementary cloud API recommendations. We also verify the ability of LRN4HCAR to handle substitute noise and improve the visibility of long-tail cloud APIs in the recommendation results. The implementation code has made publicly available athttps://github.com/hey-mem/LRN4HCAR.

In large-scale digital service sharing scenarios, given the large number of participating users, frequent cross-domain service interactions, and high-frequency service transactions, to ensure the trustworthiness of digital services, the architecture of the digital service sharing system usually chooses blockchain as its technological foundation. This not only ensures the security and credible deposit of data, but also achieves the credible traceability of data. However, in the current blockchain-based notarization architecture, there may be potential privacy leakage during data transmission, and users are unable to choose the encryption level of their data for blockchain deposition according to their own needs. Moreover, the underlying databases in current applications using blockchain lack convenient retrieval and categorization functionalities. In this work, we propose a trustworthy blockchain solution based on permission management, which implements hierarchical encryption and enables users to flexibly encrypt data according to their needs. Additionally, through the Double Star storage system, while ensuring the reliability of the system, we have also greatly improved the efficiency of data retrieval and classification. Compared to blockchain platforms like XRP and EOS, our solution achieves superior data retrieval and classification efficiency while implementing layered encryption.

Log parsing is an essential step for log analysis, which transforms raw log messages into structured form by extracting the log templates. Automatic log parsing have been the subject of extensive research. Recently, several studies have explored improving the performance of automatic log parsing via deep-learning-based approaches, especially for the pre-trained language models. However, the use of such large-scale language models for log parsing encounters several challenges, including hallucinations, high-cost and labelling efforts. To address these challenges, this paper introduces YALP, a zero-shot log parsing solution that tackles the aforementioned challenges by utilizing the capabilities of ChatGPT in conjunction with traditional methods, without incorporating user labelling. Our experiments on 16 public log datasets shows that our method outperforms several popular traditional methods in commonly used evaluation metrics. In comparison to directly utilizing GPT for log parsing tasks, our methods demonstrates significant improvements in both efficiency and effectiveness.

The rapid increase of the data of user behaviors on the Internet brings a promising chance to better discover user preferences. Recommender systems have become a popular tool for the discovery of user preferences. One key issue is how to employ user behavior sequences to develop effective sequential recommendations, especially when behavior sequences are biased. The current sequential recommendation methods either can only mine data dependencies but ignores bias or only can learn bias but cannot mine data dependencies. To solve these problems, in this article, we propose a neural collaborative sequential learning mechanism, which learns sequential information from user behavior sequences that contain bias. We propose a neural collaborative filtering (NCF) model that fully takes advantage of all data dependencies among users, items, and biased sequential behaviors. Our sequential learning mechanism employs a self-attention mechanism to learn sequential features into an embedding space and inputs this sequential embedding into the generalized matrix factorization (GMF) model and the multilayer perceptron (MLP) model. We performed experiments on two real-world datasets and compared our model with many well-known baselines. The experimental results demonstrate that our model achieves superior performance. We also give a thorough analysis through ablation experiments and sensitivity experiments.

The multi-access edge computing (MEC) technologies have advanced rapidly, bringing the 5G network vision, particularly massive machine type communication (mMTC), closer to people. Computing tasks are offloaded to a widely distributed network edge cluster, enabling efficient and real-time sensing and interaction for mobile devices. However, limited computation and communication resources in edge devices require caution in service deployment and traffic management to maintain overall load balancing, especially during heavy network loads. We explore the performance-cost relationship and transform the optimization problem into a nonlinear integer programming problem (NIP). Our genetic algorithm-based approach, GA4CBST, outperforms baselines in efficiency and effectiveness.

Nowadays an increasing number of software developers have joined in open-source software development communities such as GitHub, and develop and share softwares in these communities. Those online communities contain a huge volume of open-source repositories. Developers commonly search from existing repositories and intend to find suitable repositories to their development requirements. However, it is time-and energy-consuming to discover suitable repositories from such a large number of candidates and it may be also hard for developers to choose accurate keywords. So an effective repository recommendation service becomes an indispensable tool for developers. There have been some solutions for repository recommendation, but existing solutions have several defects such as mediocre accuracy and ignorance of useful features. In this paper, we develop a new personalized repository recommendation service with multi-modal features learning. We propose to mine two modes of features and jointly utilize the mined multimodal features. One of the features is the developers’ sequential behavior features and the other is text features of repositories. We design novel features learning mechanisms for the two modes of features. We performed sufficient experiments on a real-world dataset and the experimental results demonstrate that our model generates superior recommendation results and produces an improvement of 15.3% and 14.5% in Precision and Recall compared to well-known existing methods.

Abstract Malware detection is an important task for the ecosystem of mobile applications (APPs), especially for the Android ecosystem, and is vital to guarantee the user experience of Android APPs. There have been some exiting methods trying to solve the problem of malware detection, but the methods suffer from several defects, such as high time complexity and mediocre accuracy, which seriously decrease the practicability of existing methods. To solve these problems, in this study, we propose a novel Android malware detection framework, where we contribute an efficient Application Programming Interface (API) call sequences extraction algorithm and an investigation of different types of classifiers. In API call sequences extraction, we propose an algorithm for transforming the function call graph from a multigraph into a directed simple graph, which successfully avoids the unnecessary repetitive path searching. We also propose a pruning search, which further reduces the number of paths to be searched. Our algorithm greatly reduces the time complexity. We generate the transition matrix as classification features and investigate three types of machine learning classifiers to complete the malware detection task. The experiments are performed on real‐world Android Packages (APKs), and the results demonstrate that our method significantly reduces the running time and produces high detection accuracy.

Abstract Due to its development agility, continuous delivery, scalability and other characteristics, the microservice architecture systems (MASs) have provided complex business functions to hundreds of millions of users in many application fields. The operation and maintenance governance for a large number of microservices with complex relationships is crucial to ensuring the stability and reliability of an MAS. Although this research field has received certain attention and produced some innovative results, there is a lack of systematic reviews covering the different aspects of it. In this context, the central objective of this study is to carry out a systematic literature review (SLR) in this field, in an attempt to review existing issues, discuss the main trends, and share the findings with the academia. As a result, we start from more than 500 scientific papers published from 2009 to 2021 and extract 144 most significant papers, identify that the main research directions of this field include load balancing, fault detection, and autoscaling. Subsequently, we provide a comprehensive description of these research directions, discuss them in particular detail. We also determine limitations of current work and discuss new directions worth exploring in the future. Consequently, the outcomes will assist professionals and experts in the industry as well as academic researchers to focus more on operation and maintenance governance of MASs and further improve the relevant methods and theoretical systems in this field.

Sentiment analysis is an important task in social computing and behavior analysis, and is a typical indicator of social health. It is a challenging mission to predict the sentiment of people in socially implemented Internet of Medical Things (IoMT) systems. The existing methods have several defects, and a typical defect is that most methods ignore the fact that there is much noise in IoMT systems and it is far not enough only to develop classification models for one type of data. In socially implemented IoMT systems, many methods treat the review text as plain text but ignore the potential knowledge structure. To solve those problems, in this article, we propose a novel solution, which is composed of adversarial learning and a hierarchical attention mechanism. We construct a hierarchical attention mechanism to learn the knowledge structure of a text. We propose to apply the attention mechanism both at the word level and sentence level, enabling us to learn the knowledge from each word and each sentence. We propose to use adversarial learning to learn new knowledge as non-random perturbations, which promotes the model’s robustness. We evaluate our method on several large-scale real-world datasets, covering a wide range of cases of sentiment analysis. Experimental results demonstrate that our method achieves superior performance compared to state-of-the-art methods.

The research on Wi-Fi sensing has been thriving over the past decade but the process has not been smooth. Three barriers always hamper the research: 1) unknown baseband design and its influence; 2) inadequate hardware; and 3) the lack of versatile and flexible measurement software. This article tries to eliminate these barriers through the following work.First, we present an in-depth study of the baseband design of the Qualcomm Atheros AR9300 (QCA9300) NIC. We identify a missing item of the existing channel state information (CSI) model, namely, the CSI distortion, and identify the baseband filter as its origin. We also propose a distortion removal method.Second, we reintroduce both the QCA9300 and software-defined radio (SDR) as powerful hardware for research. For the QCA9300, we unlock the arbitrary tuning of both the carrier frequency and bandwidth. For SDR, we develop a high-performance software implementation of the 802.11a/g/n/ac/ax baseband, allowing users to fully control the baseband and access the complete physical-layer information.Third, we release the PicoScenes software, which supports concurrent CSI measurement from multiple QCA9300, Intel Wireless Link (IWL5300), and SDR hardware. PicoScenes features rich low-level controls, packet injection, and software baseband implementation. It also allows users to develop their own measurement plugins.Finally, we report state-of-the-art results in the extensive evaluations of the PicoScenes system, such as the >2-GHz available spectrum on the QCA9300, concurrent CSI measurement, and up to 40 and 1 kHz CSI measurement rates achieved by the QCA9300 and SDR. PicoScenes is available athttps://ps.zpj.io.

Vehicular ad hoc networks (VANETs) provide self-organized wireless multihop transmission, where nodes cooperate with each other to support data communication. However, malicious nodes may intercept or discard data packets, which might interfere with the transmission process and cause privacy leakage. We consider historical interaction data of nodes as an important factor of trust. Thus, this paper focuses on the trust node management of VANETs, which aims to quantify node credibility as an assessment method and avoid assigning malicious nodes. First, the integrated trust of each node is proposed, which consists of the direct trust and the recommended trust. The former is dynamically computed by historical interaction records and Bayesian inference considering penalty factors. The latter defines trust by third-party nodes and their reputation. Second, the process of trust calculation and data communication calls for timeliness. Therefore, we introduce a time sliding window and time decay function to ensure that the latest interaction information has a higher weight. We can sensitively identify malicious nodes and make quick responses. Finally, the experimental results demonstrate that our proposed method outperforms bassline methods, especially with respect to the packet delivery ratio and security.

With the development of communication technology and artificial intelligence of things (AIoT), transportation systems have become much smarter than ever before. However, the volume of vehicles and traffic flows have rapidly increased. Optimizing and improving urban traffic signal control is a potential way to relieve traffic congestion. In general, traffic signal control is a sequential decision process that conforms to the characteristics of reinforcement learning, in which an agent constantly interacts with its environment, thus providing strategy for optimizing behavior in accordance with feedback in response. In this paper, we propose multiagent reinforcement learning for traffic signals (MARL4TS) to support the control and deployment of traffic signals. First, information on traffic flows and multiple intersections is formalized as input environments for performing reinforcement learning. Second, we design a new reward function to continuously select the most appropriate strategy as control during multiagent learning to track actions for traffic signals. Finally, we use a supporting tool, Simulation of Urban MObility (SUMO), to simulate the proposed traffic signal control process and compare it with other methods. The experimental results show that our proposed MARL4TS method is superior to the baselines. In particular, our method can reduce vehicle delay.

Abstract The multiturn dialogue system has been prevalently used in e‐commerce websites and modern information systems, which significantly improves the efficiency of problem solving and further promotes the service quality. In a multiturn dialogue system, the problem of intention classification is a core task, as the intention of a customer is the basis of subsequent problems handling. However, traditional related methods are unsuitable for the classification of multiturn dialogues. Because traditional methods do not distinguish the importance of each sentence and concatenate all sentences in the text, which is likely to generate a model with low prediction accuracy. In this paper, we propose a method of multiturn dialogue classification based on key sentences mining. We design a keywords extraction algorithm, mining key sentences from the dialogue text. We propose an algorithm finishing the computation of the weights of each sentence. According to the sentence weight and the sentence vector, the dialogue text is transformed to a dialogue vector. The dialogue text is classified by a classifier, and the input is the dialogue vector. We conducted sufficient experiments on a real‐world dataset, evaluating the performance of the proposed method. The experimental results show that our method outperforms the related methods on a series of evaluation metrics.

Abstract Sentiment classification is a key task in sentiment analysis, reviews mining, and other text mining applications. Various models have been proposed to build sentiment classifiers, but the classification performances of some existing methods are not good enough. Meanwhile, as a subproblem of sentiment classification, positive and unlabeled learning (PU learning) problem widely exists in real‐world cases, but it has not been given enough attention. In this article, we aim to solve the two problems in one framework. We first build a model for traditional sentiment classification based on adversarial learning, attention mechanism, and long short‐term memory (LSTM) network. We further propose an enhanced adversarial learning method to tackle PU learning problem. We conducted extensive experiments in three real‐world datasets. The experimental results demonstrate that our models outperform the compared methods in both traditional sentiment classification problem and PU learning problem. Furthermore, we study the effect of our models on word embedding. Finally, we report and discuss the sensitivity of our models to parameters.

With the prevalence of web techniques and Internet-of-Things networks, an increasing number of developers build software by invoking existing application programming interfaces (APIs), especially in industrial systems. As the number of existing APIs in industrial systems is large, it is critical to recommend suitable APIs from big APIs data to developers in industrial software development. There have been some approaches proposed for APIs recommendation, but the existing approaches focus on the utilization of historical invocation records but ignore the exploitation of other information in the development process. We find that this ignored information can be mined as cognitive knowledge to learn the behavior rules of developers. In this article, we propose a holistic personalized recommendation framework that contains two individual models and one ensemble model, which are based on joint matrix factorization and cognitive knowledge mining. In the two individual models, we study the hidden relationships among users, which are mined from the APIs following records. We also study the hidden relationships among APIs, which are mined from the content information. We also propose an ensemble model. We crawled a large real-word dataset and conducted sufficient experiments, and compared our framework with well-known existing methods. The experimental results demonstrate that our framework achieves the best performance.

Abstract Due to the positive impact of ride sharing on urban traffic and environment, it has attracted a lot of research attention recently. However, most existing researches focused on the profit maximization or the itinerary minimization of drivers, only rare work has covered on adjustable price function and matching algorithm for the batch requests. In this paper, we propose a request matching algorithm and an adjustable price function that benefits drivers as well as passengers. Our request-matching algorithm consists of an exact search algorithm and a group search algorithm. The exact search algorithm consists of three steps. The first step is to prune some invalid groups according to the total number of passengers and the capacity of vehicles. The second step is to filter out all candidate groups according to the compatibility of requests in same group. The third step is to obtain the most profitable group by the adjustable price function, and recommend the most profitable group to drivers. In order to enhance the efficiency of the exact search algorithm, we further design an improved group search algorithm based on the idea of original simulated annealing. Extensive experimental results show that our method can improve the income of drivers, and reduce the expense of passengers. Meanwhile, ride sharing can also keep the utilization rate of seats 80%, driving distance is reduced by 30%.

With the prevalent application of Internet of Things (IoT) in real world, services have become a widely used means of providing configurable resources. As the number of services is large and is also increasing fast, it is an inevitable mission to determine the suitability of a service to a user. Two typical tasks are needed, which are service recommendation and service selection. The prediction for Quality of Service (QoS) is an important way to accomplish the two tasks, and there have been a series of methods proposed to predict QoS values. However, few methods have been used to study the QoS prediction in IoT environments, where contextual information is vital. In this article, we develop a holistic framework to attack the QoS prediction in the IoT environment, which is based on neural collaborative filtering (NCF) and fuzzy clustering. We design a fuzzy clustering algorithm that is capable of clustering contextual information and then propose a new combined similarity computation method. Next, a new NCF model is designed that can leverage local and global features. Sufficient experiments are implemented on two real-world data sets, and the experimental results verify the effectiveness of the proposed framework.

Web service recommendation is one of the key problems in service computing, especially in the case of a large number of service candidates. The QoS (quality of service) values are usually leveraged to recommend services that best satisfy a user’s demand. There are many existing methods using collaborative filtering (CF) to predict QoS missing values, but very limited works can leverage the network location information in the user side and service side. In real-world service invocation scenario, the network location of a user or a service makes great impact on QoS. In this paper, we propose a novel collaborative recommendation framework containing three novel prediction models, which are based on two techniques, i.e. matrix factorization (MF) and network location-aware neighbor selection. We first propose two individual models that have the capability of using the user and service information, respectively. Then we propose a unified model that combines the results of the two individual models. We conduct sufficient experiments on a real-world dataset. The experimental results demonstrate that our models achieve higher prediction accuracy than baseline models, and are not sensitive to the parameters.

In this paper, we present the design of JTang HSS (JTang Healthcare Service Platform for the Senior), a healthcare service platform for the senior based on cloud computing technology. In our system, services provided by third parties such as hospitals, physical examination centers, communities and some other institutions, can be integrated, managed and optimized seamlessly. The platform contains three subsystems, which are health services bus platform (HSB), Middleware Platform (MP) and health services library (HSL). HSB is constructed to facilitate data access for third parties. Meanwhile, we build the health data center to recommend services in a personalized manner based on middleware platform. Further, external services and platform owned services are orchestrated and integrated to a health services library, which supports an end-to-end process and supplies application services for the senior. The goal of the platform is to provide overall management and sufficient services in all aspects of healthcare of the senior. In detail, application services mainly include data collecting, health status assessing, entertainment supplying, social interaction, etc. Most of the services are provided as mobile applications running in devices like mobile phones and tablet computers.