Personalized Recommender System Using A Social Network Based Collaborative Filtering Technique

Point of interest (POI) recommendation has attracted a lot of research attention by combining user ratings and POIs to find the similarity of users to help them to locate an enjoyable place. However, social networks, which have become a part of modern lifestyle, contain much information about the relationship between users and POIs, such as checkin activity. POI recommendations should consider the significant features of check-in data from location-based social networks (LBSNs) for more precise results for POI recommendations. This paper presents a personalized recommender system using a social network based collaborative filtering technique that recommends the top-n POIs to the user. The proposed method calculates similarity based on user ratings with a collaborative filtering technique and user check-in activity on a social network to make personalized recommendations. We used a real-world dataset from Foursquare to test our method. The results from experiments demonstrate that using social network check-in activity combined with a collaborative filtering method can increase the performance of the recommender system in the real word.