Enterprise credit evaluation method based on gray fuzzy

Abstract

Disclosed is an enterprise credit evaluation method based on gray fuzzy, which includes the following steps: initializing multi-dimension time series data; dividing credit evaluation grade standards, and confirming various credit evaluation indexes; mapping values of various credit evaluation indexes to a certain value interval by utilizing simple mathematical function transformation in a same credit index system; confirming reference sequences and comparison sequences, calculating gray correlation coefficient, and calculating gray correlation degree to obtain gray incidence matrix composed of credit evaluation values; and converting the gray incidence matrix into fuzzy similar matrix, subjecting the fuzzy similar matrix to square self-synthesis method to be converted into fuzzy equivalent matrix, selecting a confidence level value lambda belonging to a range from 0 to 1, and calculating lambda stage matrix of the fuzzy equivalent matrix. When the rij is less than or equal to the lambda, a sample xi and a sample zj can be combined into a same class and the obtained classification is an equivalent classification on the lambda level, accordingly different evaluation results are achieved. The enterprise credit evaluation method has the advantages of reducing the calculating complexity, having good timeliness, and effectively improving the reliability.

Description

technical field [0001] The invention relates to the field of enterprise credit evaluation, in particular to an enterprise credit evaluation method suitable for processing large-capacity multi-dimensional real-time data of finance and taxation. Background technique [0002] Financial real-time data usually has the characteristics of large quantity, high data traffic burst, and high dimension. How to deal with such huge real-time data and how to use such huge data to conduct enterprise credit evaluation has always been a difficult problem. Because these real-time data have high dimension, high noise, high redundancy, and high data mutation, if such data is used for enterprise credit evaluation, the computational complexity will be high, and the timeliness will be greatly reduced. At the same time, if such data is used to evaluate the credit of enterprises, the result will inevitably be in danger of being distorted. [0003] Because we collect corporate financial data in real ...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the shortcomings of existing enterprise credit evaluation methods, such as high computational complexity, poor timeliness, and poor reliability, the present invention provides a gray fuzzy-based enterprise system that reduces computational complexity, has good timeliness, and effectively improves reliability. credit evaluation method

Images