Context-Aware Recommendation Based On Review Mining

Negar Hariri, Bamshad Mobasher, Robin Burke, Yong Zheng

Abstract: Recommender systems (RS) play an important role in many areas including targeted advertising, personalized marketing and information retrieval. In recent years, the importance of contextual information has attracted many researchers to focus on designing systems that produce personalized recommendations in accordance with available contextual information of users. Comparing with traditional RS which mainly utilize users’ rating history, the context# aware recommender systems (CARS) can result in better performance in various applications. In this paper, we present a context#aware recommender system that extracts contextual information from a textual description of user current situation and use it in combination with user ratings history to compute a utility function over the set of items. The item utility shows how much it is preferable regarding user current context. In our system, the context inference is modeled as a supervised topic#modeling problem in which the set of categories for a contextual attribute constitutes the topic set. As an example application, we used our method to mine hidden contextual data from customers' reviews for hotels and use it to produce context#aware recommendations. Our evaluations suggest that our system can help produce better recommendations in comparison to a standard kNN recommender system.