Social Internet of Things (SIoT) recommender systems are designed to enhance the functionality and efficiency of the Internet of Things (IoT) by incorporating social networking principles. They can recommend services, devices, or actions based on the social friendships and interactions between IoT devices and their users. Today, edge-driven, Multi-Agent Deep Reinforcement Learning (MADRL)-based recommender systems offer significant advantages for optimizing friendship paths and improving SIoT network navigability. Utilizing edge computing , these systems process cached services in proximity to their source, thereby improving real-time decision-making and decreasing latency. The edge-driven aspect distributes computational load , reducing central server dependency and enhancing scalability and resilience. On the other hand, MADRL enables adaptive learning from complex SIoT user interactions and network dynamics, ensuring contextually relevant and personalized recommendations. This approach improves network navigability by dynamically optimizing routes and efficiently utilizing network resources through distributed learning . Overall, these recommender systems provide scalable, adaptive, and efficient recommendations, fostering stronger social connections and improving the functionality of SIoT ecosystems. In this article, we have developed an innovative edge-driven, MADRL-based SIoT recommender framework that surpasses the performance of the leading baselines. By leveraging the decentralized processing power of edge computing and sophisticated MADRL-oriented algorithms, the suggested framework significantly enhances the optimization of SIoT friendship paths and network navigability. Extensive experimental results demonstrate that our approach achieves superior accuracy, efficiency, and scalability compared to existing state-of-the-art methods, thereby offering more personalized and contextually relevant service recommendations while reducing delay and improving real-time decision-making in dynamic SIoT environments.

The issue of security in fog has several challenges and one of which is to create a secure channel that connects fog layer to the end devices. Therefore, efficient and secure key distribution is very important to create this channel. One solution is key pre-distribute that the keys generation and distribution between nodes are before the network establishment. This article presents an algorithm for key pre-distribution in fog networks that focuses on the relationship between end devices and the fog layer. The proposed scheme is a hybrid scheme that uses the K-MEANS algorithm for nodes hierarchical clustering. Then the residual design is used to generate the key chain and BIBD design is used for key pre-distribution in headers. The proposed method is completely independent of the mobility model of and its communication and computation overheads are of the order of O(q). Also, the network connectivity is above 0.8 and close to the BIBD design, which is a fully connected KPS, and the probability of capturing the entire network in case of capturing the nodes is almost equal to 0. In the proposed method for a network with size N, the size of the key chain is N4.

Belief mining and the study of public opinion provide valuable information. Analyzing the feelings and belief mining of social media data leads to understanding users' opinions and has wide applications in decision making and policymaking. This article applies a new method based on deep learning to solve the problems of belief mining for Persian comments on the Twitter. In this method, first, the data is preprocessed with a deep neural network and then classified into political, cultural, economic, and sports classes, and the sentimental polarity is obtained. SentiPers is applied on four different datasets from Persian Twitter, Digikala store, Google translator, and synonym for evaluation. Then the results are compared with other machine learning and deep learning methods such as neural network, support vector machine, DNN, CNN, and LSTM. Python software has been used to implement this method. The accuracy of the proposed word embedding method for LSTM, CNN, DNN on the SentiPres dataset is 0.931, 0.923, 0.916 respectively. For the TF-IDF method, it is 0.837, 0.863, 0.883 respectively. that the accuracy of LSTM-WSD, CNN-WSD model has increased by 8% and 6% compared to TF-IDF. The results show that the LSTM and Word embedding methods work best.

Summary Security in fog computing is multi‐faceted, and one particular challenge is establishing a secure communication channel between fog nodes and end devices. This emphasizes the importance of designing efficient and secret key distribution scheme to facilitate fog nodes and end devices to establish secure communication channels. Existing secure key distribution schemes designed for hierarchical networks may be deployable in fog computing, but they incur high computational and communication overheads and thus consume significant memory. In this paper, we propose a novel hierarchical key pre‐distribution scheme based on “Residual Design” for fog networks. The proposed key distribution scheme is designed to minimize storage overhead and memory consumption while increasing network scalability. The scheme is also designed to be secure against node capture attacks. We also demonstrate that, in an equal‐size network, our scheme reduces node storage overhead significantly. Our research paves the way for building an efficient key management framework for secure communication within the hierarchical network of fog nodes and end devices.

Development of semantic web and social network is undeniable in the Internet world these days. Widespread nature of semantic web has been very challenging to assess the trust in this field. In recent years, extensive researches have been done to estimate the trust of semantic web. Since trust of semantic web is a multidimensional problem, in this paper, we used parameters of social network authority, the value of pages links authority and semantic authority to assess the trust. Due to the large space of semantic network, we considered the problem scope to the clusters of semantic subnetworks and obtained the trust of each cluster elements as local and calculated the trust of outside resources according to their local trusts and trust of clusters to each other. According to the experimental result, the proposed method shows more than 79% Fscore that is about 11.9% in average more than Eigen, Tidal and centralised trust methods. Mean of error in this proposed method is 12.936, that is 9.75% in average less than Eigen and Tidal trust methods.

Abstract Wireless sensor networks have been established for a wide range of applications in adversarial environments, which makes secure communication between sensor nodes a challenging issue. To achieve high level of security, each pair of nodes must share a secret key in order to communicate with each other. Because of the random deployment of sensors, a set of keys must be pre‐distributed, so that each sensor node is assigned a set of keys from a key pool before the deployment. The keys stored in each node must be carefully selected to increase the probability of key share between two neighboring nodes. In this paper, we consider a hybrid key pre‐distribution scheme based on the balanced incomplete block design. We present a new approach for choosing key pool in the hybrid symmetric design that improves the connectivity and scalability of the network. We also introduce an extension to the proposed approach to detract memory usage and improve resilience. Experimental results verify the performance and applicability of our approach. Copyright © 2014 John Wiley & Sons, Ltd.

Abstract Dynamic wireless sensor network (DWSN) is a group of mobile sensor nodes deployed in an intended area. Secure communication in DWSNs depends on the existence of an efficient key management scheme. Because of the movements of sensor nodes and unknown mobility pattern, dynamic key management is an important issue in such networks. In this paper, we propose a new key management scheme, which uses key pre‐distribution and post‐deployment key establishment mechanisms for DWSNs. The proposed approach ensures that the two communicating nodes share at least one common key. It also provides efficient ways for key generation and revocation as well as addition or deletion of mobile sensor nodes. Compared with the other key management schemes, which are designed for DWSNs, our simulation and analytical results demonstrate the efficiency of the proposed approach in terms of confidentiality, resilience, memory usage, energy consumption and overhead. Copyright © 2014 John Wiley & Sons, Ltd.

ABSTRACT Combinatorial designs are powerful mathematical tools with comprehensive and simple algebraic structures. Recently, many researchers have used combinatorial designs as key predistribution schemes in wireless sensor networks (WSNs). Previous studies on security of WSNs are mainly concentrated on those networks containing nodes with the same capabilities. Further, investigations reveal that high reliability and lifetime on networks can be achieved through hierarchical heterogeneous wireless sensor networks, where a small number of sensor nodes have more energy, memory, and transmission capability. Inspired by scheme due to Lee and Stinson, we propose a key predistribution scheme for a clustered heterogeneous WSN using transversal designs. This proposed scheme assigns key chains to sensor nodes before deployment and separates key pool of each cluster by adding a pseudo‐random generated number after deployment. The performance evaluation and security analysis show that our proposed scheme can provide better security with significant reductions on communication overhead and storage space than other key management schemes without compromising connectivity. Copyright © 2014 John Wiley & Sons, Ltd.

An important issue in database (DB) systems is responding to different users’ queries in an acceptable time. To do this, we should define different queries based on users’ real needs and we should consider suitable solutions. In this article, we express a new query called ‘boundary query’ which is used for achieving an overall view of a subject in the DB. This query does not return all query answers but it returns boundary values that cover all answers for the related query. In this article, we map a DB environment to a vector space based on necessary attributes. Then we implement the proposed method, and based on the results, we observe that the proposed method's run time is acceptable for huge DBs.