Crm System, Customer Value And Customer Segmentation Study

Propelled by the evolution of management theories and the advancement of information technology, CRM (Customer Relationship Management) has become one of the hottest topics and trends for all enterprises around the world in the 21st century. In the theories of CRM, customer value and customer segmentation play the pivotal role of supporting the customer centric management philosophy, and has recognized by the business.Customer Value is one of the key control of CRM. The ultimate purpose of Customer Management is to create more Customer Value. The evaluation of customer value can help the enterprise begin the management of customer value. The evaluation outcome can help the enterprise find new profit-making point and direct the management and purchase of enterprise.Customer segmentation gives us better understanding of customers. The main difference between modern customer segmentation model and traditional customer classified method has concentrated on different segmentation variables and on ways. Traditional customer classified method mostly select some basic aspects of customer as its income , age, value etc. Such tradition classified method is too simple. The modern customer management needs advanced researches to customers.Starting from the myriad of characteristics of customer value and segmentation in CRM, this thesis studies the data mining theories and methods relevant to customer value and segmentation. The research results presented in this thesis can serve as the foundation for new supporting technologies and methodologies for CRM.This thesis firstly performs a thorough structural analysis of the customer value problem in CRM. Based on the researches,some new analysis ways are proposed. Secondly, the thesis analyzes the various characterist-ics of customer segmentation, and points out that because of the time variations, uncertainties, currently available computing ways are not adequate. To mend those inadequacies, this thesis proposes two classific -ation models based on data mining theories. Using Probability Statistic Bootstrap way find main customer segmentation evaluation index in rapidly analyzing lots of classifying modulus. The efficiency of this way is better than the artificial neural network classification model etc. This paper researches into using multiple classifiers fusion model to integrate different results of other basic classifiers and proposes a new way of integrating.Lastly, based on the work outlined above, a prototype model of customer segmentation is designed. This prototype model' s main functions include data retrieval, data pre-processing, modeling and model application.This thesis puts forward some ways to solve the critical problems and deficiencies of the existing data mining when applying to customer value and segmentation. Its results perhaps will contribute to future applicat -ions of data mining to CRM.