In recent years, global digitalization has developed rapidly. Driven by the gradual maturity of various technologies, the popularization of Internet and mobile devices, the Internet of Things and cloud computing services, the growth of various data has greatly increased and diversified data. The value of these data can be used to predict consumer behavior, differentiate user groups, develop effective marketing strategies, and create dif-ferentiated competitiveness. To predict consumer behavior in purchasing insurance products, this study collected 4,474 insurance transactions from a bank in Tainan, Taiwan. Af-ter data preprocessing, the number of available transactions is 3,430. In these organized transactions, we use the classification of the insurance product as the dependent variable and the features of the customer as the independent variable. Then, correlation analysis was performed by chi-square test, and uncorrelated factors were analyzed. Analyze influencing factors through a decision tree machine learning model. According to the analysis results of the decision tree model, the accuracy rate is almost 70%, and the most important influencing factors are the actual insurance fee and currency. These two influencing factors can be used as a reference for the precise marketing of Tainan Bank in Taiwan.