Complementary Product Selection in E-Commerce

Abstract

Item recommendation is considered as an important feature for e-commerce sites. Item recommendation can be categorized into alternative recommendation and complementary recommendation. Alternative item recommendation technologies are quite mature and widely adopted. However, complementary item recommendation is rarely explored although most people consider this type of recommendation very important. To the best of our knowledge, this work is the pioneer in the area of complementary recommendation. Our prototype yields very high item coverage so we can generate recommendations for most of our products. Further, our system also yields fairly good precision, i.e. items recommended are deemed relevent by editors.