The implementation of knowledge-based recommender system for electronic commerce using Java expert system library

Abstract

This paper deals the design and implementation of product recommender system on a Internet shopping mall. In e-commerce application, sometimes potential buyers may be interested in receiving recommendations about what to purchase. The mainstream of an automated recommender system is collaborative filtering. In this paper, we present a knowledge-based product recommender system. The knowledge base of product domain and inference engine is implemented by JESS (Java expert system shell) and Java servlet. The system gathers the user's requirements on a particular product by questioning the user and consulting its knowledge base to find the items that best meet the user's requirement. For the recommender system, the design of product knowledge base are discussed. As a sample implementation, a cellular phone recommender system is built where the system tries to find a phone model that best meets the user's requirements.