Personalized Multi-Faceted Trust Modeling in Social Networks

In this work we develop a comprehensive approach for multi-faceted trust modeling (MFTM) and use it to model trustworthiness of peers in online social networks. We then show how this data-driven design supports clustering-based personalization of trust prediction, yielding better performance than competing solutions when applied to Yelp datasets for an item recommendation task. In all we demonstrate the promise of personalized trust modeling in social networks.