CPERS: Contextual and Personalized Event Recommender System

Abstract

This paper proposes CPERS, a contextual and personalized event recommender system that exploits overall user preference and context influences to produce recommendations in event-based social networks (EBSNs). Diversely from items in traditional recommendation scenarios (e.g. movies, songs), events in EBSNs are only valid for a short period of time, having no explicit feedback. Therefore the event recommendation problem is essentially cold-start. To overcome this limitation, CPERS combines content preferences and context influences derived from users' historical events. In particular, besides content preference based on events' description, CPERS exploits temporal impact from users' time preference, spatial constraints based upon geographical preference, cost consideration derived from expenditure history and social influence from social relationship between hosts and users. Furthermore, CPERS integrates the above factors to rank events for personalized recommendation. We collect a real-world dataset from a popular EBSNs called "Douban Events", and the experimental results on the dataset demonstrate that CPERS improves recommendation performance.