Personal loan marketing is a critical decision for a commercial bank's development of its consumer finance business in Taiwan because this business comprises majority of the bank's revenues. Efficiently and effectively reaching customers who have a high level of intention to borrow money is an important goal of banks in such marketing campaigns. The purpose of this research is to assist a commercial bank in developing a marketing model for estimating customers' intention to apply for personal loans from a market segment of customers who has already used the other banks' revolving credit of credit cards and are thus considered as potential customers for personal loans. Data mining techniques, including logistic regression, decision tree, neural networks, and support vector machines, are adopted in the model development. This research yields some interesting findings and demonstrates the effectiveness and efficiency of data mining in developing target marketing models for commercial banks.