Implementation of a Machine Learning-based Trust Management System in Social Internet of Things

Abstract

IoT (Internet of Things) is a constantly expanding network made up of billions of connected physical things, including many sensors, embedded devices, smartphones, and wearables. These actual things are typically referred to as “smart objects”. A development of the Internet of Things, the Social Internet of Things (SIoT) combines ideas of social networking to create social networks of linked smart items. They sift through the SIoT looking for services and relevant information. In the early stages of research, the concept of trust and trustworthy in social communities created by SIoT is still relatively new. We outline the foundations of SIoT and the ideas of trust in SIoT while highlighting the parallels and discrepancies between IoT and SIoT. In order to determine an aggregate trust score, all of the trust features are aggregated using a machine learning-based algorithm. Results show that the suggested trust-based model effectively separates the network’s trustworthy and untrustworthy nodes.