Development of a reliable trust management model in social internet of things

In this paper, we introduce the first application of the belief propagation algorithm in the design and evaluation of trust management systems with an introduction of a novel paradigm of social internet of things (SIoT), as ‘social network of intelligent objects’, that are based on the notion of social relationships among objects. Further, we address a trust-making process, where a person needs to make a judgement about the trustworthiness of another community member where they do not have any prior knowledge about each other. Our proposed model uses various performance measures such as: centrality and transitivity measures for SIoT analysis and then employs hybrid fuzzy nearest neighbour with Bayesian belief network and Bayesian belief network with conditional independence to represent a trust-based evaluation. Bayesian belief propagation technique is used here to infer trustworthiness in a social context. Finally, we perform non-parametric Friedman two tail test for statistical significance of the results obtained for various approaches. The evaluation of the model is done on datasets collected from epinion.com and slashdog.org shows promising results that enable us to steer the interaction among the billions of objects which will crowd the future IoT.