A Novel Hybrid Approach based on Profile Expansion Technique to Improve Cold Start Problem in Recommender Systems

Due to the growing volume of information available on the Web, the data search process is performed with spending a lot of time. In order to avoid wasting time of users, recommender systems provide information for them which is likely to be useful and valuable. Collaborative filtering is the most popular approach to provide recommendations for the users in recommender systems. However, it suffers from some problems such as cold start problem. In this paper, we present a novel hybrid approach based on profile expansion technique to improve the cold start problem in the recommender systems. In the proposed method, we take into consideration user’s demographic data beside user’s rating data in order to find an enrich neighborhood set for the active user. The results of experiments on MovieLens dataset showed that the proposed method outperformed the other recommendation methods.