In this paper, we study the problem of personalized restaurant recommendations. Specifically, we develop a probabilistic factor analysis framework, named RMSQ-MF, which has the ability in exploiting multi-source information, such as the users' task, their friends' preferences, and human mobility patterns, for personalized restaurant recommendations. The rationale of this work is motivated by two observations. First, people's preferences can be affected by their friends. Second, human mobility patterns can reflect the popularity of restaurants to a certain degree. Finally, empirical studies on real-world data demonstrate that the proposed method outperforms benchmark methods with a significant margin.

Microblogging recommendation is to find useful microblogging messages that users are really interested in. Existing methods of microblogging recommendation have low personalization for the reason of focusing on the whole microblogging social network. On microblogging site, different users have different active degree, and different users have considerable variability in structures of followees or followers. It is therefore most necessary to understand microblogging from an individual user's perspective. In this paper, we propose a novel model of microblogging recommendation based on ego networks, called EgoRec, which has the ability of capturing personal interests more accurately. Specifically, the key idea of EgoRec is to make recommendation by homophily of ego networks. We take egos as target users being recommended, while alters as source users for recommendation. Homophily between individual users can be measured by ego profile factors and ego-alter social relations. We define three main social relations in ego networks: support relation, affinity relation and focal relation for recommendation. Moreover, by detecting latent factor and combining with the homophily factors in ego networks, it makes EgoRec more personalized. Extensive experiments on real-world data sets clearly validate the effectiveness and efficiency of the EgoRec model.