ABSTRACT

Currently, economic conditions in Taiwan are facing a downturn. The ratio of nonperforming loans in financial institutions is too high. How to lower these ratios and liquidate the nonperforming assets are important issues for financial institutions. In this study, the decision tree algorithm is used to construct a model for acquiring bad accounts’ taxation information for financial institutions. It is based on ‘The essential of acceptance for creditor inquiring about the taxation information for debtor’, which was released by the Financial Information Center of the Ministry of Finance on 2013.06.19, by importing this data into a database. The total amount of book credit is NT$25.6 billion. We construct through the well-known ‘Decision Tree’ in the field of ‘Data Mining’. The results show that acquiring bad accounts’ taxation information for financial institutions can be achieved quickly and efficiently through the decision tree algorithm. By using this method, financial institutions are able to select those samples that are suitable for inquiry and reach efficient levels of management.