With the rise of Network Function Virtualization, services in Mobile Edge Computing (MEC) networks can be abstracted as Service Function Chains (SFCs), which consist of an ordered sequence of Virtual Network Functions (VNFs). To provide reliable SFC-enabled services, it is common to deploy shareable backup VNF instances to protect against primary instance failures. However, the unbounded backup sharing inevitably leads to backup contention during simultaneous failures, which will increase service recovery time. To address this issue, we formulate theJoint optimization of resoUrce consumPtion and recovery tIme under probabilisTic protEction foRSFCBackuP(JUPITER-BP) in MEC networks. In this optimization, the recovery time is integrated with the expected waiting time caused by backup contention. Further, the probabilistic protection constrains the maximum number of primary instances that each backup can protect. This constraint is aimed to balance the resource consumption and recovery time. To solve JUPITER-BP, we design a Dynamic Programming-based Greedy Backup Deployment (DP-GBD) algorithm with a provable approximation ratio. Simulation results demonstrate that the proposed DP-GBD outperforms baselines in terms of resource consumption, recovery time, and running time, while satisfying reliability requirements.

With the widespread application of edge computing, service providers cache resources on edge servers to meet users’ low latency requirements. Due to resource constraints, how to match candidate service sets deployed on edge servers has become a research hotspot. Most traditional service matching methods ignore group preferences in edge scenarios, which leads to poor service matching performance. Besides, these methods use simple embedding vectors or service usage frequencies to represent service, which lack deep mining of service representation. To improve the performance of service matching, this paper proposes a Service Matching method based on Group Preference and Service Representation Learning (SM-GPSRL). Group preference representation is obtained through deep embedding clustering and recurrent neural network. Graph attention network is adopted to compute the interaction between services for learning representation of services. And service matching is implemented based on the obtained representation. Experimental results on real-world datasets demonstrate that SM-GPSRL has a better performance compared with baseline methods.

With the massive growth of edge devices, how to provide users with video recommendation services in a mobile edge environment has become a research hotspot. Most traditional video recommendation methods regard the relationship between user and neighbor to be linear and ignore higher-order connectivity among users, which results in poor recommendation performance. Besides, these methods use a single feature to represent user preferences, which cannot effectively alleviate the data sparsity problem. To improve the performance of video recommendation, this article proposes a multi-feature video recommendation method based on hypergraph convolution (MVRHC). Hypergraph convolution is adopted to compute user neighborhood-level features for modeling high-order correlations among users. Final features are obtained by fusing multi-party features through attention mechanism. And video recommendation is then implemented based on the obtained features. Experimental results on two real-world datasets demonstrate that MVRHC has better performance compared with baseline methods.

With the massive growth of Internet-of-Things (IoT) devices, how to provide users with recommendation services in the IoT environment has become a research hotspot. Group discovery, as a prerequisite step of group recommendation that can be used to assist groups of users to select services in IoT-enriched environments, has an important impact on recommendation performance. However, existing group recommendation solutions assume that a user belongs to a specific group and ignore the possible correlation between the user’s preferences and other groups’ preferences. In addition, existing solutions treat group members as equal individuals and assign them equal weights, which makes it hard to meet the user’s accurate recommendation requirements. Furthermore, these methods focus on group members’ explicit preference information while ignoring implicit preferences. To address these problems, we propose a group discovery method based on collaborative filtering and knowledge graph (GD-CFKG). This method first uses the attention mechanism to learn the embedding of service entities from knowledge graphs and interaction between users and services to achieve users’ own preferences embedding. Considering that the preferences of similar users will help to attain accurate target user’s preferences, we then train users’ final preferences embedding by collaborative filtering and word2vec method. We conduct experiments to evaluate our approach using the MovieLens and Douban data sets. Experimental results show that our proposed method has better group recommendation performance than those baseline methods.

Mobile edge computing (MEC), which merits in reducing response time by executing services in close proximity to end devices, has recently emerged as one of promising solutions for mobile services and hence attracted increasing research. Edge nodes usually pre-download parts of private data which are stored in the cloud to enable end devices' fast access, which mitigates overburden of the centralized cloud. However, malicious attackers or unreliable services providers may corrupt with those private data on edge servers. To this end, how to validate data integrity in MEC environment safely and efficiently has become a crucial problem. In this paper, we propose a ZSS Signature based Data Integrity Verification solution for MEC (ZSDIV-MEC), which supports privacy protection and public auditing by introducing a third party auditor (TPA) and the employment of ZSS signature scheme. In the proposed ZSDIV-MEC, system architecture is described followed with a detailed verification protocol, which takes full consideration of verification for data on three cases including single edge, multiple edges, and a joint of multiple edges and the center cloud. Performance analysis of the protocol including feasibility, security, privacy, and dynamicity is discussed respectively. Experimental results in comparison with baseline schemes demonstrate that our solution has better performance in data verification and computation overhead.

Sequential recommendation system has received widespread attention due to its good performance in solving data overload. However, most of the sequential recommendation methods assume that user’s preferences only depend on specific items in the current sequence and do not consider user’s implicit interests. In addition, most of the previous works mainly focus on exploiting relationships between items in the sequence and seldom consider quantifying the degree of preferences for items implied by user’s different behaviors. In order to address these above two problems, we propose an implicit preference‐aware sequential recommendation method based on knowledge graph (IPAKG). Firstly, this method introduces knowledge graph to exploit user’s implicit preference representations. Secondly, we integrate recurrent neural network and attention mechanism to capture user’s evolving interests and relationships between different items in the sequence. Thirdly, we introduce the concept of behavior intensity and design a behavior activation unit to exploit the degree of preferences for items implied by a user’s different behaviors. Through the activation unit, the user’s preferences on different items are further quantified. Finally, we conduct experiments on an Amazon electronics dataset and Tmall dataset to evaluate the performance of our method. Experimental results demonstrate that our proposed method has better performance than those baseline methods.

Group recommendation has received great attention owing to its practical value in real applications. However, group members are implicit and groups are formed occasionally in some scenarios. Existing solutions for latent group recommendation assumes a user belongs to a specific group, and totally ignore the possible correlation between the user' s preferences and other groups' preferences. In addition, existing methods cannot deal with new items cold-start problem effectively because they only focus on which items are favored by the group without considering the hidden related information between items. These weaknesses usually lead to poor performance of latent group recommendation. To address the problems above, this paper proposes a latent group recommendation method based on double fuzzy clustering and matrix tri-factorization (DFCMTF -LGR). Firstly, this method utilizes unsupervised learning to implement potential feature extraction and double fuzzy clustering for users and items. Secondly, a novel matrix tri-factorization method is presented to adjust the membership of user-to-group, item-to-item category, and the incidence of group-to-item category is obtained. Finally, latent groups are detected according to user-to-group membership, and group rating can be generated in accordance with group-to-item category incidence matrix and item-to-item category membership. Experimental results on real datasets demonstrate that our proposed DFCMTF-LGR has better performance compared with state-of-the art methods.