INTEREST MINING FOR RECOMMENDATION SYSTEM IN VIRTUAL COMMUNITIES

Abstract

As organizations, both business and research development continue to search better ways to exploit knowledge capital accumulated on the diversified Web; it fosters the need of collaboration among people with similar interest & expertise. With the advent and proliferation of the Internet and e-commerce, it is evident that the complexity of finding relevant information on the Web has become increasingly intricate and crucial. In fact, “information overload” on the Web is a well recognized problem, where users find it increasingly difficult to locate the right information at the right time. In response to the identified need for improved users' experience by personalizing what they see and using Web 2.0 as a novel platform for users’ participation, we propose the “COMREC system” that realizes a COMmunity interest based RECommendation system. In the proposed system firstly we build an interest similarity group, an online community which is a virtual space where people who are interested in a specific topic gather and discuss in depth a variety of sub-topics related to the topic using blogs. Expert identification involves finding experts on a given topic. Thus, once the group is constructed, as our next step we identify an expert from each of the group. Expert identification in online communities is of importance as online communities can be viewed as knowledge databases where knowledge is accumulated by interactions between the members. That is, we read articles in online communities to get information on specific topics and we tend to have more confidence in the articles written by experts. On the other hand, in terms of communication dynamics, online communities are spaces where non-experts can communicate with experts and communicating with experts is not only difficult but also expensive. Consequently, in the proposed COMREC system it’s the opinion of the identified expert within a virtual community built on shared interest that constitutes the recommendation. Eventually this paradigm helps to overcome the most prominent problem existent in collaborative filtering setting, the First-Rater or the cold- start problem, as in our proposed system it is only the expert whose recommendation is considered compared to systems which require a large set of customer preferences for predicting the new preferences accurately for effective Collaborative filtering-based recommendation. The initial results show that the interest mining for recommendation system in virtual communities for building COMREC system is a motivating technique.