Personal credit risk assessment method based on IFOA-SVM

Abstract

The present invention discloses a personal credit risk assessment method based on an IFOA-SVM (Improving Fruit Fly Optimization Algorithm-Support Vector Machine). An improving fruit fly optimization algorithm is employed to perform improvement to effectively ensure the optimization precision while improving the optimization efficiency of the algorithm. The IFOA is configured to find out n elite fruit fly positions with optimal taste concentration at present and perform weighting processing of the positions to obtain an optimal weighting position so as to allow a fruit fly group to fly to the positions. The releasing operation is performed for n times to select one releasing with the optimal taste concentration mean value. The improving fruit fly optimization algorithm improves the searching probability of an optimum solution, speeds up the searching speed of the optimum solution and utilizes the optimum solution to the credit risk assessment to obtain a good assessment result, and the quantization processing of the credit assessment indexes in the sample data is performed to improve the prediction performance, have good practicability, clear out loan client credit for financing institutions such as a bank and the like and provide effective basis for reduction of the loan risk.

Description

technical field [0001] The invention relates to a credit risk assessment method, in particular to a personal credit risk assessment method based on improved fruit fly algorithm optimization SVM, and belongs to the application of artificial intelligence methods in the field of credit risk assessment. Background technique [0002] Under the situation that unsecured pure credit loans continue to heat up, all commercial banks put the loan business as the focus of development. However, in the "China Banking Industry Operation Report" released by the China Banking Regulatory Commission in 2016, the non-performing loan balance of the five major banks was 807.95 billion yuan, and the non-performing loan ratio increased by 25.7% year-on-year. The main reason that hinders the development of credit business is that commercial banks have a low level of management of loan risks and lack of effective personal credit evaluation methods. Every increase in the accuracy of the credit risk as...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to improve the fruit fly algorithm for the defect that the traditional fruit fly optimization algorithm is easy to fall into local extremum when optimizing parameters, and provide a personal credit risk assessment method for improving the fruit fly algorithm optimization SVM, effectively improving Accuracy of credit risk assessment

Images