ABSTRACT The advantages of federated learning in collaborative computing of deep learning make it a crucial approach for distributed architectures in recommender systems. However, existing federated recommender systems typically share unified item embeddings across all clients, which fails to capture user‐specific characteristics of items. How to adaptively retain both the commonality and individuality meanings of item embeddings in the recommendation becomes a critical challenge, while simultaneously preventing personalized information leakage. Therefore, this paper proposes a novel federated recommendation method (named 2P‐FedRec) that constructs a privacy‐preserving personalized recommender system in an adaptive manner. Specifically, this method employs an adaptive attention module to generate item representation containing global item embeddings (capturing cross‐user commonalities) and personalized embeddings (capturing user‐specific preferences), and utilizes two regularizers to guide the optimization of independence between these two embeddings. To protect user privacy, we also apply local differential privacy (LDP) with noise injection to the uploaded parameters, preventing the reconstruction of sensitive data. Extensive experiments on Epinions and Yelp datasets demonstrate that 2P‐FedRec outperforms the state‐of‐the‐art baselines while maintaining privacy.

The adoption of vehicular cloud computing for sequential recommendation offers flexible, reliable, and scalable computing resources in intelligent transportation systems. However, it also raises privacy concerns of drivers/passengers regarding the upload of sensitive data and models to vehicular cloud servers. To address this issue, we propose a novel privacy-preserving sequential recommendation method for intelligent transportation systems (named Grapeseed) based on split learning and variational autoencoder (VAE). Specifically, the vehicular client first inputs raw data into an encoder to produce latent variables locally and uploads these variables to the vehicular cloud server. Then, the vehicular cloud server generates and returns intermediate variables derived from these latent variables. Upon receiving these intermediate variables, the vehicular client calculates the final recommendation results. Extensive experiment results and analyses demonstrate that the proposed method improves both performance and communication efficiency between vehicular cloud servers and clients while preserving privacy.

In the context of the rapid development of service-oriented computing and cloud computing, selecting the service that meets the user’s needs from an ever-increasing number of Web services is always challenging. Exploiting auxiliary information such as a Knowledge Graph (KG) can significantly improve the effectiveness of service recommendations. However, current KG-based service recommendation methods usually merely integrate the knowledge semantic information into the user-service interaction model, which ignores the importance of global structure and does not fully consider the in-depth learning of individual user preferences. To this end, this paper proposes a Multi-Level View Contrastive Learning for Service Recommendation (MCSR) approach to address the above challenges. In particular, unlike traditional approaches that consider only two views, we consider three views, i.e., the global structure view, the local collaboration view, and the semantic view. Specifically, the user-service graph is regarded as the collaboration view, the service-entity graph as the semantic view, and the user-service-entity graph as the structural view. By applying contrastive learning across these views at different levels, MCSR fully leverages graph features and structural information, integrating auxiliary relational semantics into user-service interaction modeling. Furthermore, recognizing that the influence of auxiliary information on interactions varies between users and services, a meta-network strategy enables adaptive, personalized knowledge transfer across views, significantly improving recommendation accuracy., since the influence of auxiliary information on interactions varies between users and services, a meta-network strategy enables adaptive, personalized knowledge transfer across views, significantly enhancing recommendation accuracy. The experimental results show that MCSR significantly outperforms current state-of-the-art methods, with the positive impact of its key components on the recommended performance verified by ablation experiments.

Summary Web service recommendation as an emerging topic attracts increasing attention due to its important practical significance. As the number of available Web services continues to grow, users face the challenge of searching the most suitable services that meet their specific needs. Quality of service (QoS)‐based service recommendation becomes a popular approach to address this issue. However, existing QoS‐based service recommendation methods are inability to effectively capture valuable content and structural information from services. These methods often rely solely on low‐order explicit feature intersections in QoS information, do not fully utilize the high‐order implicit feature intersections, and ignore the rich semantic information existing in service descriptions and user preferences. To address this problem, this paper proposes a Web service recommendation method via combining knowledge distillation representation and DCNMIX quality prediction. This method combines content‐based and structure‐based service classification and service prediction based on multi‐dimensional service quality information. First, it builds a service relationship network using semantic features extracted from service descriptions. Second, it designs a graph neural network knowledge distillation framework. The teacher model extracts the knowledge of the graph neural network model, and the student model learns the structure‐based and feature‐based prior knowledge of the service relationship network. Then the student model is used to learn the knowledge of the teacher model, classify Web services, and obtain service representations. Finally, based on service representations and multi‐dimensional QoS information, it exploits the DCNMIX model to learn the explicit and implicit features intersections of Web services and obtain the prediction score and ranking of Web services. The experimental results on the ProgrammableWeb dataset show that the proposed method outperforms the state‐of‐the‐art baselines in terms of Recall, F1, Logloss, and AUC_ROC.

With the emergence of more and more Web services, finding suitable services becomes a difficult problem. Service link prediction is employed to disclose relationships among services, which facilitates the further development of service composition, selection, and recommendation. But the existing link prediction approaches simply utilize the structural features of the service network. In reality, the rich text content in service node description documents also carries latent but fine-grained semantics generated by multifaceted topic-aware factors, yet few efforts are committed to mining them. In this paper, we propose a Web service link prediction method based on a topic-aware heterogeneous graph neural network. Specifically, the method consists of two main layers, including the meta-path intra-decomposition and the meta-path inter-mergence. Meta-path intra-decomposition aims to mine the topic distribution of the meta-paths-based context while capturing fine-grained topic-aware semantics. Meta-path inter-mergence uniquely aggregates topic-aware factors according to the mined distribution and adopts a multifaceted attention mechanism to aggregate different meta-paths, enabling service nodes to generate multifaceted topic-aware embeddings that preserve not only the structure and but also the topic-aware semantics. In addition, a topic prior guidance regularization item is set up for quality assurance of multifaceted topic-aware embedding that depends on global knowledge of the unstructured text content in description documents. Experimental results on real datasets show that our proposed model outperforms other existing baselines methods in the link prediction task, successfully validating the effectiveness of our proposed method.

With the development of Internet of Things (IoT)-enabled Maritime Transportation Systems (MTS), massive data generated in the system not only requires to be stored reliably and cheaply, but also needs to be analyzed timely. The Cloud-based Maritime Transportation Systems (CMTS) allow users to upload the data without worrying about the price, capacity, location and so on. However, CMTS also brings some security issues, where the integrity protection of outsourced data is one of the most important issues since it is crucial for the safety, reliability and efficiency of sea lanes. To solve this problem, we propose an identity-based dynamic data integrity auditing scheme for CMTS. Our scheme decreases the burden of key management and improves the auditing efficiency by batch auditing. Besides, our scheme also supports dynamic operations on the outsourced data for CMTS. The security analysis shows that our scheme can ensure the feature of storage correctness and resist common attacks. In addition, the performance comparison results with other related schemes show that our scheme not only has the lowest computational cost on all entities, but also greatly reduces the communication overhead of the auditing phase. Therefore, our scheme is very suitable for data integrity verification in CMTS.

The booming Internet of Things makes smart healthcare a reality, while cloud-based medical storage systems solve the problems of large-scale storage and real-time access of medical data. The integrity of medical data outsourced in cloud-based medical storage systems has become crucial since only complete data can make a correct diagnosis, and public auditing protocol is a key technique to solve this problem. To guarantee the integrity of medical data and reduce the burden of the data owner, we propose an efficient privacy-preserving public auditing protocol for the cloud-based medical storage systems, which supports the functions of batch auditing and dynamic update of data. Detailed security analysis shows that our protocol is secure under the defined security model. In addition, we have conducted extensive performance evaluations, and the results indicate that our protocol not only remarkably reduces the computational costs of both the data owner and the third-party auditor (TPA), but also significantly improves the communication efficiency between the TPA and the cloud server. Specifically, compared with other related work, the computational cost of the TPA in our protocol is negligible and the data owner saves more than 2/3 of computational cost. In addition, as the number of challenged blocks increases, our protocol saves nearly 90% of communication overhead between the TPA and the cloud server.

In this article, we point out the security weakness of Shanget al.’s identity-based dynamic data auditing protocol for big data storage. Specifically, we identify that their protocol is vulnerable to a secret key reveal attack, i.e., the service provider (SP) can reveal the secret key of the data owner (DO) from the stored data. Further, SP can also generate a proof to pass the challenge of TPA (third party auditor) even if all block and tag pairs have been deleted. We hope that by identifying these design flaws, similar weaknesses can be avoided in future designs.

In this paper, we discuss a security weakness of Shenet al.’s public auditing protocol for cloud data [IEEE Transactions on Information Forensics and Security, 12(10): 2402-2415, 2017.]. Specifically, we point out their protocol is vulnerable to a data privacy breach attack,i.e., an adversary, once he compromises the third-party auditor latently, can also obtain all data owners’ outsourced data by constructing appropriate challenges. As a result, it breaks the property of “privacy preserving”. We hope that by identifying this design flaw, similar weaknesses can be avoided in future designs.

As the rapid development of 5G and Internet of Things (IoT) techniques, more and more mobile devices with specific sensing capabilities access to the network and large amounts of data. The traditional architecture of the cloud computing cannot satisfy the requirements, such as low latency, fast data access for IoT applications. Mobile edge computing (MEC) can solve these problems, and improve the execution efficiency of the system. In this paper, we propose a privacy preserving data aggregation scheme for MEC assisted IoT applications. In our model, there are three participants, i.e., terminal device (TD), edge server (ES), and public cloud center (PCC). The data generated by the TDs is encrypted and transmitted to the ES, then the ES aggregates the data of the TDs and submits the aggregated data to the PCC. At last, the aggregated plaintext data can be recovered by PCC through its private key. Our scheme not only guarantees data privacy of the TDs but also provides source authentication and integrity. Compared with traditional model, our scheme can save half of communication cost, and is very suitable for MEC assisted IoT applications.