Abstract
Today, most personalized and recommendation services are built around user interest extraction models but the outputs of these algorithms are ambiguous in nature. This makes it inherently difficult to understand what users are personally interested in and more importantly what they are feeling towards these interests. By studying both users' interests and emotions, simultaneously, one can further investigate the motivation behind a user's interests/intentions. Such findings can be useful to build better interest extraction models and algorithms that leverage personalized and recommendation services (e.g. ads. recommendation, professional social networks and dating sites). In this paper, we propose a new approach that uses an opinion mining technique (emotion classification) to model user personal interests on microblog data. The interests identified by our method are very consistent with a user's real and personal interests. Essentially, we analyze the contribution degrees of different positive emotions in regards to how they assist in extracting a user's interests. Our experimental results indicates that a user's emotion-bearing information can provide empirical evidence to his/her true personal interests.