The Evaluation Study Of Personalized Recommendation System On Customer Satisfaction

Personalized Recommendation System in E-commerce is a kind of advanced intelligent system based on data mining, which is used for recommending product information to customers. Personalized Recommendation System plays an important role for e-commerce website and has been adopted widely. It can provide several kinds of service to customers fast and precisely, including products recommendation, sales promotion, personalized service and cross-selling improvement. And, as it can promote customer’s loyalty and repeat purchases, customer satisfaction is important in increasing competitiveness of enterprise. Therefore, improving customer’s satisfaction of personalized recommendation system is significant to these e-commerce enterprises.After reading plentiful literature, this article makes a summary of personalized recommendation system and customer satisfaction both domestic and abroad. In this article, a customer satisfaction assessment index system for personalized recommendation system, which is based on the model of SERVQUAL, is set up, also,this assessment index system references some assessment indexes of other literature. In this assessment indicator system, there are 21 second indexes belonged to 5 first indexes, which are recommendation content, service quality, system response, technology ability and support effect. After this is make up, this article choose Analytic Hierarchy Process(AHP) and use the software MATLAB to determine the weight of every index.At last, this article analyzes customer satisfaction of personalized recommendation system of JD, Amazon and Taobao by using Grey Correlation Analysis. The data for this analysis is coming from the questionnaire and this questionnaire is based on the customer satisfaction assessment index system of personalized recommendation system. The result is that “Taobao>JD>Amazon” in customer satisfaction of personalized recommendation system. This article discusses the reason of this result by analyzing these first and second indexes further. And after this discussion, some suggests to improving customer satisfaction of personalized recommendation system is given. They are increasing the quality of recommendation content, enhancing the precision of recommendation service, strengthening the interaction between customer and personalized recommendation system, promoting the practicability and giving support for purchase decision. These will help to improving customer satisfaction of personalized recommendation system effectively.