Trust computational heuristics for social internet of things

The Internet of Things (IoT) is an evolving network of billions of interconnected physical objects, such as numerous sensors, smartphones, wearables, and embedded devices. These physical objects generally referred to as the smart objects, when deployed in the real-world, aggregate useful information from their surrounding environment. As of late, this notion of IoT has been extended to incorporate the social networking facets which have led to the promising paradigm of the Social Internet of Things (SIoT). In SIoT, the devices operate as autonomous agents and provide an exchange of information and service discovery in an intelligent manner by establishing social relationships among them with respect to their owners. Trust plays an important role in establishing trustworthy relationships among physical objects and reduces probable risks in the decision-making process. However, existing literature confronts numerous trust-related challenges, primarily due to the multi-dimensional concept of trust and a SIoT-specific trust computational model for the SIoT ecosystem. 

To address the limitations of current studies, the primary goal of this thesis is to design an efficient SIoT-specific trust computational heuristic to evaluate the trustworthiness of a SIoT object in the SIoT ecosystem. To achieve this goal, this thesis introduces the notion of trust in general and in SIoT in terms of trust definition, trust characteristics, and the current studies present in the literature. Subsequently, a trust evaluation model that employs SIoT-specific trust features, including but not limited to, SIoT relationships in terms of social characteristics (e.g., friendships, working relationships, and community-of-interest), direct observation with current and past interactions, and indirect observations as a recommendation from credible SIoT objects, is proposed to augment the idea of trust quantification by integrating all the trust features via conventional weighted sum approach to get the final trust score of a SIoT object. Furthermore, a number of trust aggregation approaches (e.g., conventional weighted sum, machine learning, and artificial neural networks) are envisaged in order to aggregate independent trust and address the challenges of the conventional aggregation approach.

Finally, to prove the validity of the proposed heuristics, extensive experiments are conducted in order to analyze and evaluate the performance in terms of accurately classifying the trustworthy vs. non-trustworthy SIoT objects, monitoring the dynamic behaviour of objects, and comparison with the state-of-the-art vis-à-vis different real world datasets.