Trustable user recommendation using semantic provenance model

Abstract

Online Social Network (OSN) has become one of the social media tool to connect with old friends, meet new interesting people and create relationships irrespective of geographical locations. Advancement of telecommunication and mobile technology easily paved the way for the faster community growth in OSN. Community in OSN grows through friend suggestion and friend requests. Current system of friend suggestion is based on the existence of link structure between mutual friend and Friend of A Mutual Friend (FOAMF) rather than evaluating intimacy and interaction behavior. The friend requester may be an imposter. The research work proposes the semantic provenance model to identify the imposters and provide highly relevant user recommendation. The current research work is aimed at developing a trustable user recommendation system based on semantic provenance model to provide potential friend suggestion and friend request. Provenance data of users for trust computation are extracted from the timeline and chat sessions of Facebook data. Implicit user interactions such as comment, chat and like and explicit content features are used to compute trust. newline