An Indirect Social Trust Model for Vehicular Social Networks Using Evolving Graph Theory

TL;DR

This paper introduces a novel indirect social trust model for vehicular social networks using evolving graph theory and cryptography, addressing trust propagation challenges in highly dynamic environments.

Contribution

It presents a new trust model that accounts for the dynamic nature of VSNs and employs cryptographic techniques to handle subjectivity in trust estimation.

Findings

Model effectively captures trust evolution over time.

Cryptographic approach ensures privacy and accuracy.

Computational analysis shows model efficiency.

Abstract

The increasing importance and consequent challenges of establishing indirect trusted relationships in highly dynamic social networks such as vehicular social networks (VSNs), are investigated in this paper. VSNs are mobile social networks that aim to create social links among travellers on the roads. Besides matching interests between two users, social trust is essential to successfully establish and nurture a social relationship. However, the unique characteristics of VSNs pose many challenges such as uncertainty, subjectivity and intransitivity to indirect social trust modelling. Furthermore, the current trust models in the literature inadequately address trust propagation in VSNs. We propose a novel indirect social trust model for VSNs using evolving graph theory and the Paillier cryptosystem. We consider the VSN as a highly dynamic social evolving graph where social ties among vehicles hold a trustworthiness factor that evolves over time. This factor is estimated based on the behaviours, opinions, distances, and communication metrics of the parties involved. Employing the homomorphic property of the Paillier cryptosystem, the proposed model targets the subjectivity problem when combining multiple opinions to establish an indirect trusted relationship. Through analysis of computational and communication complexities, we show the viability of the proposed model and the efficiency of its indirect trust computation algorithm.