Study On Personalized Recommendation Method In E-Commerce Based On Similar Cases Analysis

The development of the E-commerce gives consumers more and more products to choice, at the same time, information overload becomes increasingly serious. It is not easy for customers to find what they need, even though they are submerged in huge amounts of information. In this case, the personalized recommendation E-commerce comes into being. The personalized recommendation in E-commerce, by simulating the behavior of sales staff, provides product information and advice for Consumers through E-commerce websites to help consumers to decide what to buy and complete the purchase process. To consumers, the personalized recommendation in E-commerce helps them find what they interest in the massive information. To merchants, the personalized recommendation in E-commerce helps them implement diversity management, improve cross-selling, bring huge economic benefits. So far, the study of personalized recommendation problems in E-commerce has attracted the attention of business and scholars. It has appeared some research results. But informed research paid less attention to the classification of the consumers and the preference analysis of different types of customers. Furthermore, the level of products quality and service in the sales process is the two important factors for consumers to make decisions whether to purchase. Therefore, further study will be needed. It is of great important theoretic and practical significance.The informed research has shown that similar cases analysis method meets customers thinking characteristics, recommended result can easily accepted by customers. So it is necessary to introduce similar cases analysis method into E-commerce personalized recommendation and consider the classification of the consumers as well as the level of products quality and the service. The purpose of this thesis is to make a deep theoretic analysis and methodological research on of personalized recommendation method in E-commerce based on similar cases analysis. A series of research works are conducted as follows.(1) The problem of personalized recommendation in E-commerce based on similar cases analysis is described and the theoretical framework for the research problem is proposed. In order to have a specific analysis of customer information, the research problem is classified into two categories according to the customer that if they have demands when logging the website, i.e., the personalized recommendation in E-commerce considering the potential demand of customers and the personalized recommendation in E-commerce considering the given demand of customers. In addition, the research problem is described, and the research framework for solving the problem is proposed. These basic research works mentioned above provide a theoretical guidance framework and analytical frame work to the related problems and lay a solid foundation to extended issues and applications.(2) The study of personalized recommendation method in E-commerce considering the potential demand of customers is conducted. Specifically, in view of the potential demand of customers, a personalized recommendation method in E-commerce based on similar cases analysis is proposed. First, use the register information, behavior information and evaluate information to analyze the preference of such customer, and propose the case representation as well as the problem description considering the potential demand of customers. Then, give the corresponding similarity calculation method based on customers have purchasing behavior whether or not. Finally, determine the set of recommended products by incorporating the level of products quality and service.(3) The study of personalized recommendation method in E-commerce considering the given demand of customers is conducted. Specifically, in view of the given demand of customers, a personalized recommendation method in E-commerce based on similar cases analysis is proposed. First, use the demand information, behavior information and evaluate information to analyze the preference of such customer, and propose the case representations well as problem description considering the given demand of customers. Then, give the corresponding similarity calculation method based on customers have purchasing behavior whether or not. Finally, determine the set of recommended products by incorporating the level of products quality and service.The proposed personalized recommendation method in E-commerce based on similar cases analysis considers the classification of consumers, the level of products’ quality and service. It provides a reference method to the personalized recommendation problem in E-commerce.