Design And Implementation Of Analytical CRM Based On Association Rule Mining Algorithm

The object of CRM is building a system, which made the enterprise a co-ordinate and brand new entity in customer service, company competition, sale and technology endorsement, bring the persistent competitive advantage. the CRM system is a typical interaction system which is the total sum of management method of integration information technology and application solutions, it is the series information technology, method and process in helping enterprise manage customer relation, also is the software and hardware system for enterprise sale, customer service and workflow automation.Enterprise has accumulated large quantity transaction data, because they have built information building for many years. they are eager to draw the potential knowledge from the abundant transaction data. So the analytical CRM are researched by many research institutes.This paper is composed of six chapters, In first chapter, giving the project research process and the object according to CRM definition and the CRM researching domestic and foreign situation. In the second chapter, the classification of CRM are discussed, build the ordinary CRM model and the analytical CRM model. In the third chapter, database and data mining are discussed, build the CRM model based on data mining in the foundation on analyzing the Fayyad model and CRISP-DM model, and design the process of Aprior algorithm.In the fourth chapter, analyzing the component technology of J2EE technology standard, and build the analytical CRM frame based on J2EE technology.In the fifth chapter, providing the design of analytical CRM based on Struts frame applying with MVC model and give the functional requirement, build the analytical CRM system model in details using UML. Design the whole system dataflow and build the CRM database. Finally, give the implementation of CRM relying on struts frame. In analytical model block, drawing the association rule between product and sale region from sale transaction data by applying with the algorithm of Aprior.