Fuzzy prototype model and semantic distance

摘要

It is a challenge to provide an intelligent product suggestion for these new customers without previous shopping records in the supermarket application. To solve such a problem, we design a hybrid fuzzy expert system for recommendation using the improved fuzzy prototype model and semantic distance. Moreover, we implement a demonstration of the recommendation system by using intelligent Fril/SQL interrogator, which is an object-oriented and knowledge-based support query system containing the set of reusable logic objects linking one another. The results are evaluated by comparing the average product frequency of recommendations with the average frequency of non-recommendations.