Mining customer product ratings for personalized marketing

With the increasing popularity of Internet commerce, a wealth of information about the customers can now be readily acquired on-line. An important example is the customers' preference ratings for the various products offered by the company. Successful mining of these ratings can thus allow the company's direct marketing campaigns to provide automatic product recommendations. In general, these recommender systems are based on two complementary techniques. Content-based systems match customer interests with information about the products, while collaborative systems utilize preference ratings from the other customers. In this paper, we address some issues faced by these systems, and study how recent machine learning algorithms, namely the support vector machine and the latent class model, can be used to alleviate these problems.