Incorporating evidence into trust propagation models using Markov Random Fields

Current trust models for social networks commonly rely on explicit voting mechanisms where individuals vote for each other as a form of trust statement. However, there is a wealth of information about individuals beyond trust voting in emerging web based social networks. Incorporating sources of evidence into trust models for social networks has not been studied to date. We explore a trust model for social networks based on Markov Random Fields, which we call MRFTrust, that allows us to incorporate sources of evidence. To allow comparative evaluation, a state-of-the-art local trust algorithm-MoleTrust-is also investigated. Experimental results of the algorithms reveal that our trust algorithm that incorporates evidence performs better in terms of coverage. It is competitive with the MoleTrust algorithm in prediction accuracy and superior when focusing on controversial users.