BP neural network credit assessment method based on principal component analysis method

Abstract

The invention relates to a BP neural network credit assessment method based on a principal component analysis method. According to the BP neural network credit assessment method based on the principal component analysis method, government data related to an individual is combed out of bank data, sample data is formed in combination with a credit assessment result of a bank on the individual, and the prediction performance is improved after the sample data is subjected to unitization processing; dimensionality reduction is carried out on the sample data with the principal component analysis method, complex indexes and multi-dimensional data types can be handled, the demand for processing big data is better met, the credit assessment result of the bank on the individual serves as a reference for training a BP neural network model, and thus a credit assessment model based on government big data is established; the subjectivity of expert scoring can be avoided, credit inquiry can be provided for enterprises or individuals, the credit system of financing institutions is also supplemented, and high classification accuracy and practicability and a good assessment result are achieved.

Description

technical field [0001] The invention relates to a BP neural network credit evaluation method based on principal component analysis method. Background technique [0002] At present, credit evaluation research is mainly carried out in financial institutions. Based on the business data collected by the institutions themselves, the credit reports of enterprises and individuals are obtained through the analysis and evaluation of professionals. Relying solely on business data within financial institutions for credit evaluation can easily lead to one-sided conclusions. When financial institutions face customers with less information, they often cannot draw valuable credit assessments. [0003] With the advent of big data, multi-faceted data fusion analysis has become the mainstream, especially government big data plays a pivotal role in credit reporting. The Chinese government holds more than 80% of the data, but due to various limitations and the influence of departmental intere...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a BP neural network credit evaluation method based on principal component analysis, which can overcome the subjectivity of expert scoring, has high classification accuracy, practicability, and good evaluation effects, and can solve complex problems. indicators, multi-dimensional data types, more in line with the needs of big data processing

Images