Study And Application Of Data Mining Technology On Life Insurance Business

Data mining is a new technology of KDD that is developed for business data statistic and analysis. In this paper, the theory and implement of data mining are discussed based on marketing database of life insurance company. At first, the development of technology and characteristic of data mining are introduced, its function and task in insurance business application are also explained. After several mature algorithms and software tools are compared, and a data mining process carried out by Clementine of SPSS is presented. In this process, clustering methods for customer behavior analysis and mining association rules for marketing research are applied. According the CRISP-DM process model, customer records in Comprehensive Business Processing System are randomly extracted. Then the efficiency of K-means, BIRCH and Kohonen models in clustering and Apriori, GRI and Carma algorithms in mining association rules are studied. Finally, some helpful result for the decision-making and merchandise planning are concluded.