Unifying Geographical Influence in Recommender Systems via Matrix Factorization

In recent years, we have witnessed the development of location-based services where geographical information plays an important role in reflecting user preferences. This paper aims to provide a unified framework for location-aware recommender systems with the consideration of geographical influence using the matrix factorization method. In the framework, we propose three models corresponding to three kinds of ratings, namely, ILARS-MF to non-spatial ratings for spatial items, ULARS-MF to spatial ratings for non-spatial items and UILARS-MF to spatial ratings for spatial items. The experimental results on real data sets show that our recommendations are more effective than baseline methods.