Research On Customer Value Classification Based On Rough Set And Neural Network

The development of the Internet technology have a great effect on business activitiesin many respects. E-commerce, as a new field of application of Internet technology, haspenetrated into all aspects of traditional business activities, which has a profound impact onthe modern way of life. Customer is the source and profits of the enterprise survival, tomore and more enterprises are from the commodity oriented business strategy is customeroriented, therefore the research of customer value to the development of the era ofe-commerce enterprises has important significance. And e-commerce trading platform canbe convenient for the customer access behavior, purchase behavior were recorded, theoperators of enterprises have also begun to focus more on customer behavior analysis,which can be effectively divided based on customer, help enterprises to find the value of thecustomer, to improve efficiency and benefit, to seize the initiative in the competition.Traditional customer classification method and evaluation system of customer value oftraditional have their own limitations, based on the summary on their advantages andshortcomings, and combined with the current popular data mining techniques, theclassification system of the value evaluation of rough set neural network based clients, is awhole set of treatment solutions for customer classification problem based on customervalue. Through the research on the problem of customer value classification problems andcustomer, and discusses the optimal fusion method of rough set theory and neural networktechnology in this field, and the large number of technical details of the problem proposedin the evaluation system is studied and discussed.Finally, through the example of the classification and evaluation of value to thecustomer a mobile communication industry, a set of methods using the results from dataacquisition to the final classification evaluation is proposed in this paper by experiment,and achieved the desired results, verifies the effectiveness of the proposed evaluationsystem.