Credit Risk Of Individual Housing Loans Based On Data Mining Technology Assessment Study

The paper deals mostly with large data of individual housing loan by some data mining technologies. Through uncovering the hiding models in the large data, this paper finally gets a credit risk assessment model.These studies have four parts: constructing a credit risk assessment system^ data selection, data processing and constructing model by data mining tools. These four parts are all based on some relevant theories of data mining and the generic process of data mining. Among these four parts, constructing a credit risk assessment factors system is the basic of the others, data selection and data processing is prepared for mining, and constructing models is the result of research. So these four parts are closed connected, and they are all necessary steps of data mining.According to the requests of commercial banks, contents that the thesis studies can be shown in the following aspects:1. Responding to the operations of individual housing loan and consulting expert, all factors, which are correlative with risk, are selected from the database of individual housing loan. The assessment factors system is constructed. Based on the risk assessment factors system, a data set is constructed by analyzing source file. This data set is the basic of data processing.2. Data processing is the most important step in the entire process for data mining. It decides the quality of mining in the latter. Considered the quality of data and the calculating efficiency, this paper applies appropriate processing methods to deal with the data set, and then we get a new data set that is appropriate for mining. In addition, some processing methods are also been studied and applied.3. Based on theories of data mining, we select decision tree technology to resolve the problems in credit risk assessment after comparing some mining technologies. Finally we get a mining risk assessment model, which has been tested and evaluated that has higher predictive accuracy. This model can be used to help employees to analyze every loan and can help loan department to makecorrect decision.