YXYanwei Xu

Yanwei Xu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yanwei Xu has a total of 46 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Contrastive Paradigm for Representation Learning in Interactive Recommendation

    by Jingyu Li, Zhiyong Feng, Shizhan Chen, Dongxiao He, Qinghang Gao, Hongqi Chen, Guoli Wu, Meng Xing, Yanwei Xu, Hongyue Wu

  • Collaborative-Guided Diffusion for Sequential Recommendation

    by Yanwei Xu, Ye Tian, Fangcheng Fu, Peichao Lai, Guoli Wu, Lianyong Qi, Bin Cui

  • Open Access

    Bundle Recommendation for Budget-Conscious Consumers

    by Gaoyong Han, Zhiyong Feng, Yanwei Xu

    Traditional recommender systems are dedicated to recommending items of interest to consumers, but consumers are more concerned about how to purchase optimal bundles (combinations of items) under the constraints of their limited budgets. In this paper, an approach for recommending bundles according to users’ budgets is proposed based on Graph Convolutional Networks (GCN). The GCN propagates over a heterogeneous graph consisting of the user-bundle interactions, user-item interactions, and bundle-item affiliations. The price attributes of bundles and items are also considered to capture users’ preferences at two levels: the price of a bundle as a whole and the price of individual items contained in the bundle. Besides Hit Rate (HR) and Normalized Discounted Cumulative Gain (NDCG), a metric that takes into account the similarity between a recommended bundle and the user’s target bundle is designed to evaluate the performance of our model. Experimental results show that the model achieves state-of-the-art performance on three datasets. The approach of recommending bundles according to users’ budgets is more in line with the theory of consumer choice, which provides a fresh perspective for the research on recommender systems.


  • Collaborative Trust Evaluation With Relative Attention in Evolving Online Social Networks

    by Yanwei Xu, Zhiyong Feng, Gaoyong Han, Hani Attar, Hongyue Wu, Hongqi Chen, Schahram Dustdar, Abdul Rahman Afzal

    In online social networks, trust relationships between users play a vital role in connecting like-minded friends and identifying trustworthy information. However, conventional approaches, mainly utilizing graph attention networks, often regard trust interactions as static and neglect their dynamic nature. Such a limitation undermines the precision of modeling trust dynamics and obstructs the comprehension of how trust evolves within online social networks. To address this challenge, we propose TrustFormer, a novel collaborative approach that integrates time-dependent trust interactions and explicitly predicts dynamic trust relationships. Specifically, TrustFormer first embeds an evolving topology from trust interactions to capture latent temporal features. It then applies time encoding within a multi-head attention network, quantifying the significance of time-sensitive topological features. Further, TrustFormer employs bidirectional interaction transferring to formalize shared trust topological features. Finally, pairwise trust relationships are evaluated using these topological features. Supported by extensive experiments on two real-world datasets, our approach demonstrates significant improvements in trust dynamics modeling, outperforming state-of-the-art methods in accuracy.

  • FMQS: Enabling Fast Multimedia Data Queries in Edge Storage Systems

    by Lekang Song, Qiang He, Yanwei Xu, Bing Zhao, Chao Yan, Wenwen Gong, Xiaoxiao Chi, Boyuan Yan, Weiyi Zhong, Rong Jiang

    In recent years, with the development of cyber–physical–social systems (CPSSs), the integration of the Internet of Things (IoT) and social networks has been advocated to promote further development. Multimedia big data (MMBD) generated by Social Internet of Things (SIoT) devices is commonly distributed and stored on edge servers. In this context, collaborative processing of user requests among edge servers becomes crucial to meet the low-latency retrieval needs of users. However, in practice, the lack of global data storage information leads to significant time and space costs when identifying servers across the entire system. To tackle this issue, we propose a distributed edge data indexing system named FMQS, aiming to achieve fast MMBD queries at the edge. First, we introduce the cuckoo filter (CF) tree index structure, which enables rapid edge data queries. Second, we utilize a novel indexing structure based on the grouping cuckoo bloom filter (GCF) tree to improve query performance. What is more, we conceptualize distributed edge servers as entities within social networks, optimizing their relationships in terms of geographical proximity to improve query performance. Finally, we conduct performance evaluation, security verification, and analysis of the FMQS scheme on an edge storage system consisting of 90 edge servers, further confirming the effectiveness and feasibility of FMQS in MMBD queries.

  • Fuzzy Federated Learning for Privacy-Preserving Detection of Adolescent Idiopathic Scoliosis

    by Xiaotong Wu, Yan Ding, Xiaokang Zhou, Yanwei Xu, Shoujin Wang, Xiaolong Xu, Lianyong Qi

    As a distributed intelligent paradigm, fuzzy federated learning (FuzzyFL) can reduce the uncertainty and noise of biomedical data and is suited to enhance the accurate detection of adolescent idiopathic scoliosis (AIS). The advanced paradigm requires the hospitals to share the gradient of the fuzzy deep neural network (FDNN) rather than biomedical data. Not only that, the recent research works have been devoted to privacy-preserving FuzzyFL for secure AIS detection, that adds differential privacy-based noise to the gradients against membership inference attack, attribute inference attack. However, a novel reconstruction attack called gradient leakage attack (GLA) on inferring biomedical data over the gradient brings the security challenges to FuzzyFL and thus has a negative influence on AIS detection. It is natural to ask a fundamental question: Can differentially private FuzzyFL for AIS detection over biomedical data defend GLA? In this paper, we construct a privacy-preserving FuzzyFL framework called PrivateFuzzyFL , that offers a great opportunity to present the systematic evaluation of the private FDNN threatened by the GLAs. In our experiments on a set of chest X-ray images and four FDNNs, we compare more than ten private fuzzy federated optimization algorithms in terms of the defense effect and the utility cost and derive that (1) The existing private FDNNs in FuzzyFL can offer a certain amount of privacy protection for biomedical data against the GLA; and (2) The perturbation algorithm with better defense effect usually causes the worse AIS detection of the FDNN.


  • Privacy-Aware Web APIs Recommendation for Consumer Mashup Creation Based on Iterative Quantification

    by Ruowei Zhang, Lianyong Qi, Chao Yan, Zige Chen, Wenwen Gong, Yanwei Xu, Wajid Rafique

    With the emergence of the "unmanned" field, unmanned supermarket software has entered consumers’ lives in line with the pace of development of the times. Nowadays, developers of unmanned supermarket software often tend to create feature-rich consumer mashup applications by invoking a variety of Web APIs (Application Programming Interfaces) in order to save time and costs. The number of APIs available on edge servers has significantly increased with the rise of mobile edge computing. To cope with the continuous growth in API volume and the increasing diversity of API functions in API sharing communities (e.g., ProgrammableWeb.com), lightweight recommendation techniques have been employed to assist consumer mashup developers in finding their desired Web APIs from a vast pool of candidates. However, during the data processing in edge computing, traditional Web API recommendation approaches often prioritize the accuracy of recommended APIs, neglecting the potential privacy risks associated with disclosure. This significantly reduces the willingness and incentive of consumer mashup developers to share API information, such as historical edge mashup-API invocation records. To address this issue, we propose a privacy-preserving Web API recommendation approach, WARecITQ, based on Iterative Quantification (ITQ). Specifically, we convert the sensitive mashup-API invocation records into less-sensitive mashup indices using Iterative Quantization hash coding. We then utilize these less-sensitive mashup indices as the primary decision-making criteria for Web API recommendations, thereby achieving the goal of completing top-k API recommendations. Lastly, we crawled a real-world dataset of mashup-API invocation records from ProgrammableWeb.com and conducted a series of experiments on the dataset to evaluate performance. The results of the experiments demonstrate the superiority of our WARecITQ approach in multiple performance metrics compared to other related approaches, while ensuring consumer mashup developer privacy.


  • A Smart Data-Driven Multi-Level Synchronous Digital Twin Model for Vehicle-Assisted Driving

    by Jianhang Liu, Yuxiang Wang, Wenwen Gong, Hanwen Liu, Yanwei Xu, Huaizhen Kou

    Recently, the method of non-intelligent vehicles, acting as information consumers, receiving integrated data shared to the cloud by mobile edge nodes and RSUs has garnered significant interest for its potential to enhance road safety. The method can assist the customers in achieving advanced driver assistance system functionalities. Nonetheless, challenges such as resource constraints at RSUs, long communication distances between the cloud and vehicles, and frequent data interactions make it difficult to meet the strict latency requirements in highly mobile and complex traffic environments. To address these challenges, this paper proposes a smart data driven multi-level synchronous digital twin model for vehicle-assisted driving. We make use of multi-level synchronous deduction along with high-level proactive updating to reduce update latency significantly so that it can meet the timeliness requirements for cyber-physical services in differentiated scenario. Additionally, we construct a customer matching value and dynamic vehicle model update system. It solves the resource allocation problem that arises during the update of differentiated twin models for multiple customers when resources of RSUs are constrained. While satisfying the constraints of update latency, it minimizes the cost of vehicle model updates and effectively enhance system efficiency.


  • Towards evolving software recommendation with time-sliced social and behavioral information

    by Hongqi Chen, Zhiyong Feng, Shizhan Chen, Xiao Xue, Hongyue Wu, Yingchao Sun, Yanwei Xu, Gaoyong Han