Algorithm Based On Fuzzy And Neural Network Study Of College Credit Student Loans

Credit risk is one of the most substantial risks in the operation process of commercial banks. The management of credit risk has particular sense for the security of micro- financial institutions, and plays an important role in macroeconomics and financial stability. Nevertheless, at present, there exists some problems in the choose for credit rating approaches of commercial banks, and such approaches are usually oversimplified and cannot reflect risks potently, especially in the aspect of credit rating of university student loans, the outcome of credit ratings are usually inconsistent with actual risk grade, and cannot reflect the real credit status of university student accurately.The known operation process in credit rating of university students concentrated on qualitative analysis, so in this thesis, qualitative analysis is combined with quantitative analysis, and multiple evaluation approaches are used for credit rating of individuals, such as layers analyzing approach, fuzzy approach and nerve network approach are combined to evaluate the credit rating of students precisely.This structure is arranged as follows: chapter 1 is the preface. Chapter2 contains theoretic foundation. Information asymmetry, credit information pool and game theory are introduced to analyze the construction foundation for the university student loan credit model. I give a concise introduce of enhanced multi-layer analyzing approach IAHP, fuzzy evaluation approach FUZZY, BP nerve network model NNT in this chapter. Chapter 3 is the part of reality examination, I examine and analyzed the present status of university student loan , and focus on questionnaire investigation, institution investigation and sample analysis. Chapter 4 contains the research of university student loan credit model based on fuzzy synthetic evaluation approach. Chapter 5 contains the research of university student loan dynamic credit tracing model based on BP nerve network. the outcome is convincing. The last part is conclusion.