Millions of interconnected Internet of Things (IoT) sensors and devices collect tremendous amounts of data from real-world traffic scenarios. Route planning with IoT network could derive critical value for smart city and automatic vehicles. In the current route planning methods, the route weight only decays with time and distance separately, without considering the inherent spatiotemporal dependence. Besides, lacking of fine-grained traffic data interaction analysis in dynamic environment is another challenge for route planning. In this article, we propose a spatiotemporal interactive attention neural network for personalized route planning. First, we utilize an intelligent approach to route recommendation based on data collected by IoT under given spatial constraints. Next, we carry out traffic road network analysis with the spatial graph attention structure. Then, we develop a temporal self-attention mechanism to capture multilevel temporal relationship. In particular, we explore the influence of features, such as external attributes, the spatial correlation between different locations, and the temporal correlation between different time intervals. Finally, we build an aggregation network to allocate appropriate weights to measure these features for obtaining the potential best route selection. Route planning results show that the performance of our scheme is better than that of the baseline scheme, which proves that our method makes full use of attribute information and environmental changes. The IoT experimental results demonstrated that the presented system could be advantageous for tackling IoT scenarios in a cost-effective way.

In session-based recommendation, the user's next possible click can solely be predicted based on historical interaction behavior in the ongoing session. Previously representative works mainly use sequence models and graph neural networks to model user's behaviors of the session. These works have achieved promising results, but each also has certain defects. In view of the shortcomings of the previous works, we propose a multi-encoder framework, under which the advantages of each encoder are retained. Different encoders are used to mine different session features and finally generate a more powerful session representation to improve the recommendation result. Furthermore, in order to improve the performance of recommendation, we introduce the inter-session collaboration information by designing a Inter-session Collaboration Module. Extensive experiments on two real-world datasets demonstrate the superiority of our method over state-of-the-art algorithms.

In social network analysis, link prediction is a task to predict the link possibility through the known information of the network structure. However, most current methods focus on the linear superposition of few social network attributes, which makes it difficult for relational content attributes to fully participate in the prediction. Moreover, obtaining low dimensional dense edge representation and edge weight from high-dimensional sparse social network plays a critical role in the improvement of prediction accuracy. In this paper, we propose a general framework that can predict the presence and weight of edges according to the local structure, topology and content of social networks. Firstly, we mine the representation of each node providing an exciting opportunity to advance our knowledge of feature extraction. Besides, based on the sparsity and high dimension, we use the joint embedding method to express the connection information and semantics information to learn the node representation. Furthermore, this study makes a significant contribution to research on convolutional neural network by encoding the corresponding type of node features and preserving the similarity between the original associated nodes. The prediction performance of edge presence and edge weight was experimentally investigated by large real-world datasets. The F1 index, which can measure the prediction effect of edge presence, is improved by at least 0.03. In addition, the MSE index and the PCC index of edge weight prediction are improved by at least 0.03 and 0.04 respectively. Our scheme could effectively capture the diversity of content embedding in different relational patterns.

Smart home is an indispensable part of Internet of Things(IoT) owing to the prompt development and application of smart devices. However, the data collected from smart homes usually need to be processed by a cloud server, which means there is a risk of leaking the privacy of users during the transmission. In this situation, edge computing is considered to be an ideal platform for smart home, which enable data to be processed at edge nodes. Unfortunately, because of unsecured Wi-Fi connection and smart devices, edge nodes also have the possibility to encounter malicious attacks. Hence, in this paper, we designed an intrusion detection system (IDS) to be deployed on edge nodes. We convert network traffic to images which are applied to train a convolutional neural network (CNN) to classify the categories of network traffic. Furthermore, Auxiliary Classifier Generative Adversarial Network (AC-GAN) is adopted to generate synthesized samples to expand the intrusion detection dataset. We experiment on the UNSW-NB15 dataset which contains substantial network traffic about the normal and anomalies. The proposed scheme is effective to minor categories of which precision could be improved 12%. Besides, the precision can reach 96% in binary classification about normal and anomaly.

With the rapid development of the mobile Internet and the increasing popularity of smart terminals, various mobile social applications are emerging. Medical data has become a valuable data asset and is being continuously explored and utilized, which has greatly promoted the improvement of the medical service level. However, publishing and using user data makes the user vulnerable to reasoning attacks. Due to the special nature of the medical field, medical data not only carries the health status of patients and medical process information but also involves individual sensitive information of a large number of patients. Allowing users to fully enjoy the advantages brought by social networks while ensuring security is an important issue that needs to be solved urgently in the era of big data. In this paper, we first provide an overview of social network data privacy risks and various types of attacks. Aiming at the privacy leakage of weighted social networks, we propose a privacy protection recommendation algorithm based on differential privacy. The algorithm utilizes the change of edge weight grouping, which greatly reduces the amount of calculation and satisfies the user's rapid response. It minimizes a privacy leak of user private data under data availability while supporting personalized rankings. Compared to the most advanced methods, this method protects users from reasoning attacks and reduces the distortion of ranking results caused by data confusion to ensure the accuracy of recommendations. Experiments on real-world datasets show that our framework can achieve more effective and lasting protection for user-sensitive data.

The large volume of event information makes it difficult for users to find interesting events in social networks. Therefore, we would like to develop an intelligent event recommendation to reduce information overload. Specifically, by exploring the behavior of users during the selection process, we are able to find particular rules associated with various event attributes which reflect the willingness of users. However, traditional event recommendations in social networks mainly concern the basic items like time and location. It is noted that few studies have yielded specific aspects such as the influence and spread capability of events and hosts. In this paper, we propose an event recommender approach fusing multiple features that can provide users with customized contents. To be specific, we consider hybridizing features including event influence, host impact, fee, social relationship and spatiotemporal characteristics. In order to achieve better performance, we concern the match degree between user and event properties especially in terms of their content and impact. Based on the improved idea of RankNet with neural networks, we build a Learning to Rank algorithm to reveal the importance of each feature. We rectify the problem of data sparse and cold start to grasp the balance of accuracy and novelty. Extensive experiments on datasets demonstrate that our method achieves promising results in comparison with other schemes.

This paper proposes a hybrid model for deformable template which combines alignable and non-alignable sketches. These sketches are subject to slight or considerable translations in different images. For slight translations, Wu et al proposed active basis model to capture them, where each sketch is allowed to shift in position and orientation. For larger translations of sketches, assumed that they follow the same distribution as sketches of natural image ensembles, which need not be explicitly modeled. But in fact, for a specified object class, the unaligned sketches follow a totally different distribution from those of natural images. We summarize these sketches by their means in the foreground mask. We treat the mean value in each direction as independent features and fit their marginal distributions on object ensemble and natural image ensemble using Gaussian distribution. The marginal distributions are combined with Active Basis into a joint probability ratio to distinguish foreground object from natural background. Experiments are conducted on 14 object classes, most of which show considerable improvement in ROC.

The buffered crossbar switch is a combination of crossbar-based switch architecture and combined input and output queueing scheme. It is promising to achieve good performance without complex implementation. In this paper we apply network calculus technique to study the delay performance of the buffered crossbar switch. We setup a service curve-based model for the path through which a flow traverses the buffered crossbar switch. Based on this model, we develop a technique that can determine the end-to-end service curve guaranteed to a flow and calculate the delay upper bound of this flow. Then we analyze the determining factors of this delay bound performance.