Electricity selling company customer credit prediction method and device, equipment and storage medium

Abstract

The invention discloses an electricity selling company customer credit prediction method and device, equipment and a storage medium. The method comprises the steps: obtaining customer data of an electricity selling company, and screening a customer credit data set; calculating the credit data based on an XGBoost algorithm, constructing a credit prediction model through a weak learner CART decisiontree, and grading the prediction result of the credit prediction model according to a preset credit rating; on the basis of a BP neural network algorithm, constructing a credit risk early warning model used for training prediction result historical data of the credit prediction model and outputting a next credit prediction result; and when the grade comparison result of the next credit predictionresult satisfies a preset early warning condition, outputting a credit risk early warning signal. According to the embodiment of the invention, the credit prediction model is constructed and the credit of the customer of the electricity selling company is predicted through the model, so that the electricity selling company is helped customize differentiated services for users with different credits and avoid the credit risk of the customer.

Description

technical field [0001] The present invention relates to the technical field of power systems, in particular to a customer credit prediction method, device, terminal equipment and storage medium of an electricity sales company. Background technique [0002] With the reform of the electricity market, electricity sales companies established by different capitals will become the main body of the new market, so the competition is fierce. For various electricity sales companies, how to adopt differentiated services for customers of different natures is the key means to enhance customer stickiness, and customer credit is an important nature of concern for electricity sales companies, and in the changing market environment Under the current situation, the credit situation of customers of electricity sales companies is constantly changing. How to warn customers of credit changes and take corresponding measures to prevent credit risks of customers of electricity sales companies is an ...

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a customer credit prediction method of an electricity sales company to solve the technical problem of how to give an early warning of changes in the credit of customers of an electricity sales company. Customize differentiated services for users with different credits and avoid customer credit risks

Images