Sentiment-Based DUser Interests etection By Analyzing User Profile Data

With the advent of world wide web (WWW), huge shift in the way people communicate has occurred. Today, social medias has become a major communication channel for communication between the individuals and groups. With the massive growth of data and users of these medias, by analyzing them, new horizon for knowledge discovery and users behavior pattern has opened. One of the challenge is to inferring user interests from microblogging services like twitter. Because user shared data are in textual context, extracting user requirements and interest are complex. In this paper we introduce a state-of-art Social Medias User interests extraction system that use user s profile information (tweets) to extract his/her interests by enriching them from wikipedia articles. By analyzing and modeling this information in semantic way and scoring user interests with sentiment analysis teqniques, we can present user interests profile which can be use further in recommender systems. Results show that our system performed better than the traditional systems.

Keywords:

User Interest , Social Media , Twitter , Sentiment Analysis