A Comparative Evaluation of Hybrid Product Recommendation Procedures for Web Retailers

A product recommendation is an enabling mechanism to overcome information overload occurred when shopping in an e-marketplace. Recommendation methods are a personalized information filtering technology to help customers find the products they would like to purchase. Collaborative filtering is the most successful recommendation technology but its application to e-commerce has exposed well-known limitations such as sparstity and scalability. We introduce several hybrid product recommendation procedures based on clustering, Web usage mining, collaborative filtering, and content-based filtering driven a bayesian model (CBBM) to overcome them. The recommendation quality of each hybrid product recommendation procedure is compared with others by several experimentations. Through the experiment with real Web retailer data, it is found that hybrid procedure using Web usage mining, and a bayesian model can perform recommendation tasks effectively, but using clustering analysis can perform efficiently.