Incorporating User Preferences Across Multiple Topics Into Collaborative Filtering For Personalized Merchant Recommendation

Merchant recommendation, namely recommending personalized merchants to a specific customer, has become increasingly important during the past few years especially with the prevalence of Location Based Social Networks (LBSNs). Although many existing methods attempt to address this task, most of them focus on applying the conventional recommendation algorithm (e.g. Collaborative Filtering) for merchant recommendation while ignoring harnessing the hidden information buried in the users' reviews. In fact, the information of user real preferences on various topics hidden in the reviews is very useful for personalized merchant recommendation. To this end, in this paper, we propose a graphical model by incorporating user real preferences on various topics from user reviews into collaborative filtering technique for personalized merchant recommendation. Then, we develop an optimization algorithm based on a Gaussian model to train our merchant recommendation approach. Finally, we conduct extensive experiments on two real-world datasets to demonstrate the efficiency and effectiveness of our model. The experimental results clearly show that our proposed model outperforms the state-of-the-art benchmark approaches.