Commercial Bank Based On Data Mining, Customer Credit Risk Assessment Study

Under the influence of the financial crisis, global economic has declined, and the collapse of a large number of enterprises and financial institutions has taken place, which makes commercial banking credit risk increases continuously. At the background of financial crisis, to improve credit risk management capabilities is essential. On credit risk management, currently our commercial banks even more stay in a traditional credit analysis methods and the use of rigid rules and regulations, whose knowledge, research, practice, and techniques on modern commercial bank's credit risk management are located in the superficial level. The changes of global financial competition and the rapid development in financial theory and information technology tools put forward higher requirements for credit management.This paper first proposed the study background and significance, summarized the research status of domestic and overseas, analyzed the status quo of China's credit risk assessment and focused on the point of this paper. From the theoretical basics of the commercial bank's credit risk assessment, it demonstrated the theory of credit risk assessment and data mining techniques, carried out in order, laid the foundation for the following research. And then, this paper constructed the commercial bank's credit risk evaluation index system, selected neural network which is one of data mining models to build a commercial bank credit risk assessment model, and used MATLAB and SPSS Clementine software to empirical analysis with 78 companies'financial data. This study provides a useful basis for evaluation on commercial bank's corporate customers' credit risk assessment and management, and has a certain degree of referential significance.