Research On Customer Loyalty Model Applied In E-commerce

This paper analyzes the customer’s purchasing behavior based on the e-commerce customer’s historical transaction records, we build new model for getting the CLV index, which indicates the value of customer loyalty. After data preprocessing, we can get the four CLV index:Recency, Sustainability, Visits, and Profit.Based on the RSVP model, we select the different cluster data mining algorithms to mine the customer purchasing behavior, in order to find the high loyalty customer group. K-means algorithm is the simple and effective method in partition-oriented clustering algorithms. Also this paper provides4methods to improve the methods of choosing centers in K-means algorithm.In order to verify the improved algorithm, we used R language to carry out clustering analysis and visualization the customer segmentation result. With the different data mining groups, the corresponding market advices of every customer group in different hierarchical customer loyalty is proposed.The last part of this thesis does a summary of the study, and addressed some ideas about the customer loyalty data mining in e-commerce business in the future.