Support Vector Machine Based On Fuzzy Integral For Credit Risk Model Evaluation Of Commercial Bank

As the main intermediary of financial transaction, the commercial bank is the weatherglass of state's economic condition.Commercial bank also plays a important role in reducing economical risk and the unstable factor, guarantying the national economy healthily. The commercial bank itself undertakes various types risk in the operation,including credit risk, interest rate risk, fluid risk, management risk, capital risk and policy risk and so on. Credit risk holds the special important status in each kind of risk, is the most primary factor which causes bank goes bankrupt.Credit Risk evaluation is the basic work of commercial bank's credit risk management, which goal is to analyze the credit risk of the bank in provides a loan, and support the decision of the loan. But Traditional credit risk evaluating methods always only estimate the risk factors which exist in corporate itself who wants to get debits from the assessment subject, i.e. banks, while seldom consider the influence which the bank's deposit and loan structure, credit risk condition bring upon. These methods lead to the absence of the assessment subject and results in inevitable systematic error. In this essay, I make the credit risk analysis in three respects: enterprise factors, bank's own risk condition and factor focused by new capital contract of Basel. The essay establish a commercial bank credit risk index systems that make up of 23-risk index concerned the cash-flow factor firstly,Then support vector machines and fuzzy integral are introduced. And the essay particular dissertate structure and calculation principle of support vector machines. Support vector machines ensemble has been proposed to improve classification performance recently. However, currently used fusion strategies do not evaluate the importance degree of the output of individual component SVM classifier when combining the component predictions to the final decision. A SVM ensemble method based on fuzzy integral is presented in this paper to deal with this problem. This method aggregates the outputs of separate component SVMs with importance of each component SVM, which is subjectively assigned as the nature of fuzzy logic.The result of basing on voting ensemble SVM, single SVM, fuzzy nerve...