| With the development of information system, there are many data about automobile insurance customers and sales for enterprise. Using traditional inquire or analysis tools, however, people can't recognize the useful information, let alone get decision-making for further market strategies. Data mining technology can help people to solve the very problems. Therefore, how to use data mining technology in automobile insurance CRM system becomes a hot topic.The important position of CRM was determined by the characteristics of automobile insurance. Effective CRM must be based on powerful data analysis technology. There are plentiful data produced in automobile insurance operations, which can be used to analysis, but increases greatly difficulties at the same time. How to make the plentiful data resources turn into real knowledge that can be used in business decision, which can support the automobile insurance CRM?This thesis puts forward applying data mining in automobile insurance CRM through understanding automobile insurance and business data, mainly researches on how to use decision tree and clustering technology in automobile insurance CRM system. Focal points were set on two business themes: automobile insurance risk analysis and design the corresponding data mining model, and compared this data mining model to the simply one which is using in automobile insurance; the character of insuranced automobile and design the corresponding data mining model, and compared two decision-tree algorithms.I believe that with the progress of data mining technology, insurance company will be coming to realize the importance of application of data mining in automobile insurance, and direct the insurance company to make business decision. |