Research On F E-commerce Website Customer Segmentation

Customer is the most valuable resources of Enterprise.And it is the focus of business competition.The development of e-commerce is very fast.Because the characteristics of customer mobility is convenience and low cost,the fighting for the E-commerce customer become more intense.If the enterprise want to remain invincible in the increasingly fierce competition,it have to produce or provide the differentiated and high quality products.The particularly important thing is to dig the value hiding in the huge customer consumption data.Therefore,in order to propose an effective strategy for the development of the site and maximize the allocation of resources,on the one hand the e-commerce website have to improve product quality and optimize customer service systems,but also full and effective digging out the value of customer data in the database.In this paper,the K-means clustering method in data mining is used to segment the F site customers by using the improved RFM model in combination with the customer's consumer behavior.Through analyzing the characteristics of different customer groups,make suggestions for the different customer groups and provide decision reference for the customer management.The important part of the development of the website is to segment customers and to make different marketing strategy for different different customer groups.First of all,the paper analyze customer characteristics of the F-site and the method of customer segmentation.In order to improve the method of the F-site's customer segmentation.The paper raise a analytical method that based on the RFM model of traditional customer segmentation.The new method is using profit substituting amount.Secondly,The paper divide customers into 54 categories based on Modeler software.And analysis the feature of the F-site customers.It will waste the enterprise resources because of then excessive categories.So the paper user K-means cluster to classify customers.And it devide F-site customers into five different catagories.Finally,the paper analysis the different feature between catagories.Then it raise personalized suggestions for the different customer catagories.