Reseach On The Mortgage Credit Risk Evaluation On The Basis Of Decision Tree Algorithm

House loan credit risks cause concerns among the whole world due to outburst of American loan crisis in2008. According to the requests of "new Basel capital agreement " brought forward by Basel committee, the major factor to affect the banking management is not the single risk, such as financial market risk or credit risks, but the combined action results of various factors (such as:financial market risk, credit risks, operation risk, etc.). however, our national bank risk estimate is still in the junior stage, especially lack of experiences and methods on personal credit risks assessment and management comparatively. Owing to that this current situation, the paper evaluates the personal credit risk on the basis of the bank customers’data, which is practically significant in banking supervisory system perfection.This paper firstly elaborated the present research situation and the research results of the credit risk assessment in the domestic and the foreign, and simply introduced the decision tree algorithm and the mortgage credit risk. Basing on the analysis of the present research situation of mortgage credit risk assessment, this paper further analyzed the problems of the mortgage credit risk assessment. Then the article analyzed the influence factors of the mortgage credit risk, designing the corresponding evaluation index, and gave the specific steps of the evaluation index screening, and then determined the ten indicators which had major influence. Next, through the comparative analysis of the mortgage credit risk assessment methods, the paper chose the decision tree algorithm to assess the mortgage credit risk, and described the principles and evaluation model of the decision tree algorithm. On the basis of designing evaluation index and chosing evaluation method the aboved, the paper assessed the mortgage credit risk on A bank, and put forward to the strategy reducing the mortgage credit risk of A bank on the basis of the evaluation result.The decision tree algorithm the paper used can accurately evaluate the customers of the mortgage credit grade, which is not only suitable for the bank to track the credit risk assessment of the old customers, but also suitable to predict the credit rating of the new customers. The evaluation result of the decision tree algorithm can be used as the decision-making basis of which the banks provide loans to the customers, and will to play a huge role on reducing loans credit risk for the bank, and insure the healthy and stable development of the bank industry finally.