Spatio-Temporal Topic Modeling in Mobile Social Media for Location Recommendation

摘要: Mobile networks enable users to post on social media services (e.g., Twitter) from anywhere and anytime. This new phenomenon led the emergence of a line work mining behavior mobile taking into account spatio-temporal aspects their engagement with online media. In this paper, we address problem recommending right locations at time. We claim propose first comprehensive model, called STT (Spatio-Temporal Topic), capture user check-ins in single probabilistic model for location recommendation. Our proposed generative does not only captures check-ins, but also profiles users. conduct experiments real life data sets Twitter, Go Walla, Bright kite. evaluate effectiveness by evaluating accuracy The experimental results show that achieves better performance than state-of-the-art models areas recommender systems as well topic modeling.