E-commerce recommender systems (RSs) can help users quickly find what they need or new products might be interested in. To continuously enhance user trust in the website, improve page visits and dwell time, most importantly, increase gross merchandise value (GMV), it is crucial to understand capture important information hidden data, which has a great impact on choice. The fashion e-commerce websites collect attributes of items as well purchase behaviors, but lack fine-grained classification...

Abstract In unmanned aerial vehicle (UAV)-enabled fog computing networks, how to efficiently offload multiple tasks the nodes is a challenging combinatorial optimization problem. this paper, in order optimize total delay for UAV-enabled simple scheduling algorithm and multi-task offloading scheme based on fireworks (FWA) are proposed. First, system model of networks described detail. Then, proposed allocated single node. Based algorithm, all presented. Finally, simulation results show that...

Clothing images vary in style and everyone has a different understanding of style. Even with the current popular deep learning methods, it is difficult to accurately classify labels. A representation model based on neural networks called StyleNet proposed this paper. We adopt multi-task framework build make full use various types label information represent clothing finer-grained manner. Due semantic abstraction image labels fashion field, using simple migration method cannot fully meet...

For D2D-based vehicular communication networks, how to effectively and quickly allocate frequency resources is a big challenge. In this paper, resource allocation scheme based on immune algorithm for networks proposed. Firstly, the system model described. Then issue in formulated analyzed. Next, an efficient presented. Finally, analysis simulation results validate high efficiency of proposed compared with other algorithms. This result can provide guidance deployment VUE system.

In this paper, the resource allocation problem for device-to-device (D2D) communications underlaying cellular networks is formulated and analyzed. our scenario, we consider that number of D2D user equipment (DUE) pairs far larger than equipments (CUEs). Meanwhile, blocks are divided into two types: CUEs ones DUEs. Firstly, system model presented, formulated. Then, a scheme based on genetic algorithm proposed. To overcome dedicated not fully shared in algorithm, an improved harmony search...

It is of great significance to study the robustness different networks under attacks. To attack network more effectively, a new centrality strategy for path based on multi-hop by mapping problems into percolation, which called centrality, proposed in this paper. This method tested respectively two kinds model (ER network, SF network) and some real removing with largest from iteratively, then calculating efficiency, maximum size connected sub-graph corresponding number The experiment results...

To prevent information leakage and enhance the authentication security for intelligent terminals, this paper proposes an method based on Jensen-Shannon divergence, which uses behavior characteristic distribution probability vector(BCDPV) obtained by normalizing reconstructing behavioral characteristics as aided beyond passwords. The of proposed was evaluated results are presented using False Rejection Rate (FRR), Acceptance (FAR) Equal Error Rate(ERR), shows that do best job identity...

Using unmanned aerial vehicles (UAV) as the wireless base station is effective to provide service ground users (GU) compared conventional communications. In this paper, an efficient algorithm proposed for deployment optimization that improves throughput UAV communication. First, communication system model presented includes a certain number of uniformly distributed GUs and some UAVs problem get larger described. Second, details optimal beetle antennae search (BAS) are presented. Third,...

Abstract In UAV-enabled fog Computing networks, how to efficiently offload multiple tasks the computing nodes is a challenge combinatorial optimization problem. this paper, in order optimize total delay for UVA-Enabled Fog simple scheduling algorithm and multi-task offloading scheme based on fireworks (FWA) are proposed. First, system model of networks described detail. Then, proposed allocated single node. Based algorithm, all presented. Finally, simulation results show that performance...

The macro nature of social networks has always been a hot topic scientific research, which is great help in studying the characteristics human behavior. spectrum closely related to network and determines certain information, including community division, loop, diameter, degree so on. To ensure security data, differential privacy framework used propose three algorithms for protection edge weights query. Firstly, basic concepts weight neighbor graph difference are given. For single singular...