Based Crm And Data Warehouse Implementation,

Competetion between corporations has changed from product to customer. With the development and the extending of the chain of the supply, demonds on the management of the end customers are now becoming more and more important. Customer becomes one of the most important resources of enterprises. Efficient management for customers has already become the urgent requirement of enterprises. Hence, customer relationship management (CRM) emerges with the tide of the times. Based on data warehouse technology, this paper discusses the technology to construct the CRM application system in insurance field, as well as the development steps, technology difficulties and the ways of how to carry out the Application.CRM system is applied to insurance company more and more widely. Making use of data warehouse technology, insurance companies improve the their information levels to make better use of their data , to implement decision-support system (DSS) and to maintain their company's competitive edge. The operating data of insurance industry is different greatly from others', because its historical data is modified frequently. So building the insurance CRM system based on data warehouse is a very hard project.First, this paper outlines the origin, ingredients (make-up), developing conditions of CRM, the requirement in insurance field and the important work that author had done in actual project. Second, the paper introduces the concept and trait of CRM, data warehouse and OLAP system. Then, for the most part, this paper introduces the construction of all parts of MTU-MENULIFE CRM and the main technology it uses. The main develop tools of this paper are oracle data warehouse and oracle database. The display tools this paper use are Brio Explorer. The data source of this paper is oracle database of menu-life insurance company 's OLTP system. This paper talks about the main technology that MTU-MENULIFE CRM system uses, as well as the current system of menulife insurance company.Since the work of ETL accounts most part of data warehouse project, the paper introduces a lot about the implement of ETL. At the same time, this paper makes out a way making use of association rule to solve the problem of insurance policy's inspection. To be specific, First, to get the support and the confidence making use of machine learning. Second, to inspect the risk of the new policy making use of confidence. In another words, the rule that we use to inspect new policy is deduced from claim data before. In addition, this paper makes some efforts in the way of performance tunning of data warehouse.The quality of customer's data is of vital importance to the success of data warehouse project. First, this paper introduce a common way on cleaning data of data warehouse. Second, this paper introduces a way based on classic algorithm to clean the data of CRM. This algorithm including data preprocess, data aggregration in advance, and reducing the number of matching improves the performance of eliminating duplication of customer data. But the matching of the classic algorithm serves for the custom of English. So we also make out a way to match the customerCRMdata according to custom of Chinese. At last, the paper modifies the process of matching of customer data by the way of adding one more time to scan the customer data. The modification of classic algorithm involves two points. One is adding one more time to scan the customer's duplication data and reducing the length of the window this algorithm uses. The other is making use of more keys to match the customer's data. In practice, this modification of the classic algorithm can improve the efficiency greatly and reduce a large amount of matching time. This algorithm can not only be used to inspect duplication of customer's information, but also can be applied to inspect the cheat of insurance policy. So it is very useful to avoid risk of policy and to improve the efficiency of customer service for insurance company. And it can also be used in other fields such as credit loan decision supports of ban...