Applying fuzzy decision tree to infer abnormal accessing of insurance customer data

Insurance has become an important way of investment, savings and risk management. Abnormal accessing of customer data has recognized a big problem to cause the loss of insurance companies' benefits and reputation. In this paper, we analyze the behavior of abnormal accessing from the operational records of operators in order to detect the possibilities of the abuse of customer data. An algorithm of fuzzy decision tree was used to classify the categories of operators' behavior and conduct the report of abnormal accessing. It offers decision-making support for the related management. After testing experiments, our approach is more effective and efficient than the current approach which manually determines the abnormal behavior by employees.