NeuralWalk: Trust Assessment in Online Social Networks with Neural Networks

Abstract

Assessing trust in online social networks (OSNs) is critical for many applications such as online marketing and network security. It is a challenging problem, however, due to the difficulties of handling complex social network topologies and conducting accurate assessment in these topologies. To address these challenges, we model trust by proposing the three-valued subjective logic (3VSL) model. 3VSL properly models the uncertainties that exist in trust, thus is able to compute trust in arbitrary graphs. We theoretically prove the capability of 3VSL based on the Dirichlet-Categorical (DC) distribution and its correctness in arbitrary OSN topologies. Based on the 3VSL model, we further design the AssessTrust (AT) algorithm to accurately compute the trust between any two users connected in an OSN. We validate 3VSL against two real-world OSN datasets: Advogato and Pretty Good Privacy (PGP). Experimental results indicate that 3VSL can accurately model the trust between any pair of indirectly connected users in the Advogato and PGP.

TrustGNN: Graph Neural Network based Trust Evaluation via Learnable Propagative and Composable Nature

“…We adopt two widely-used, real-world datasets for model evaluation (as done in most related works such as Guardian [23], NeuralWalk [13] and OpinionWalk [9]). The statistics of the two datasets are shown in Table II.…”

“…The probability statistics-based methods [10]- [12], rely on statistical distributions to represent and model social trust relationship in a computational way. The machine learning-based methods [3], [13], use some machine learning techniques such as matrix factorization to model the trust evaluation task as a learnable problem. However, these existing approaches usually have high computational complexity, or have poor performance because they do not consider the user's attribute information.…”

The 10 Research Topics in the Internet of Things

“…The importance of trust management has been recently investigated across numerous sort of networks, i.e., mobile ad hoc networks [38], peer-to-peer networks [39], social networks [40], and as-of-late for vehicular ad hoc networks within the context of the promising paradigm of Internet-of-Vehicles [41]. Evaluating trust becomes indispensable in the case of a highly dynamic and distributed network since pervasive infrastructure cannot be guaranteed at all the times in such scenarios which is imperative for public-key based cryptographic techniques.…”

A Probabilistic Graph Model for Trust Opinion Estimation in Online Social Networks

“…How to model the trust between users in OSNs has attracted much attention in recent years. Existing trust models can be categorized into four groups: topology based [3,4,5,6], PageRank based [7,8], probability based [9,10,11,12,13], and subjective logic based models [14,15,16,17,1,18,19]. Along with the rapid development of the Internet and online services, trust has been used in many applications for either improving users' quality of experience (QoE) or preventing the disturbance of malicious users.…”