A content based seller recommendation system in an open market

More and more customers are buying products on on-line stores and they will be able to make a decision to buy with ease, if they are given reliable information of sellers. But unfortunately, such information is not available and very limited at best. Thus, this paper proposes a recommendation system which recommends most dependable sellers to the customers who want to buy a product. The system first evaluates the sellers registered on an online store by classifying them either as good or as bad using a decision tree technique (J48), and selects only good sellers. Then, the system makes use of the content-based filtering method to find best-matching top-K sellers among the selected good sellers by comparing the individual seller profile and the customer profile. This study makes a contribution in that to our knowledge, it is the first attempt to recommend sellers to customers, not products as is done in other studies.