Combining social media and location-based services for shop type recommendation
Miao Tian, Zhiwen Yu, Zhu Wang, Bin Guo · 2015
It is an important yet challenging task for investors to determine the most suitable type of shop (e.g., restaurant, fashion, etc.) for a newly opened store. In this paper, we present a shop type recommendation system, which can be used for multiple parties to make investment decisions. We adopt two types of features, location features and commercial features to model a shop, which are derived from the heterogeneous data, i.e., social media and location-based services (LBS). A novel bias learning matrix factorization method with feature fusion is proposed for recommending an appropriate shop type for a given location. Experimental results show that the proposed method outperforms state-of-the-art solutions.