Credit risk prediction model construction method and system, storage medium and electronic equipment

Abstract

The invention relates to a credit risk prediction model construction method and system, a storage medium and electronic equipment. The credit risk prediction model construction method comprises the steps: constructing a feature library which comprises a plurality of feature variables for predicting the credit risk of an enterprise; constructing a training sample set according to multiple predefined risk types, wherein training samples comprise black samples and white samples; based on the feature library and the training sample set, training each risk type to obtain a corresponding predictionmodel; and fusing the obtained multiple prediction models to obtain the enterprise credit risk prediction model. The credit risk prediction model construction method is particularly suitable for predicting the credit wind direction of small and medium-sized enterprises lacking transaction data, and the prediction model constructed by the credit risk prediction model construction method is used forpredicting the credit risk of the small and medium-sized enterprises, and the accuracy of the prediction result can be improved, and the credit risk prediction model construction method is not limited by the transaction data.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a method and system for constructing a credit risk prediction model, a storage medium and electronic equipment. Background technique [0002] Based on the era of big data, through the collection and analysis of various data generated in the operation of enterprises, it can effectively help enterprises themselves or others to create more value. For example, by analyzing the types of hot sales and buying groups of products, it can help companies make more accurate product marketing strategies. For another example, through the analysis of enterprise transaction data, it can help enterprises form credit files and help enterprises to obtain financing or loans. Taking corporate credit assessment as an example, although the traditional credit investigation model can evaluate corporate credit risk to a certain extent, the data of the traditional credit assessment model is too de...

AI Technical Summary

Problems solved by technology

Taking corporate credit assessment as an example, although the traditional credit investigation model can evaluate corporate credit risk to a certain extent, the data of the traditional credit assessment model is too dependent on transaction data, and for small and medium-sized enterprises that lack loan experience and transaction behavior, their Lack of or incomplete credit records will automatically be regarded as a high credit risk, which will affect their financing or loans

In other words, the current assessment model is not targeted and is not suitable for small and medium-sized enterprises that lack transaction data, which in turn will lead to inaccurate credit risk prediction results for small and medium-sized enterprises

Images