Automatic U-shaped binning algorithm based on quadratic function and information maximization

Abstract

The invention belongs to the technical field of credit scoring, and particularly relates to an automatic U-shaped binning algorithm based on a quadratic function and information maximization. By adopting the algorithm, an optimal binning result can be automatically generated; U-shaped binning is achieved through the characteristic variable with the U-shaped actual structure, interpretability in the actual business meaning is achieved, unnecessary operation cost consumption is reduced through optimization of the algorithm in the operation process, and finally the generated binning result givesconsideration to interpretability in the business meaning and optimal representation in the data meaning. The algorithm is embedded into establishment of a traditional score card model or other emerging machine learning, the quality of a traditional credit model in the financial industry can be obviously improved, the approval accuracy is improved, and more fraud overdue applications are refused.

Description

technical field [0001] The invention belongs to the technical field of credit scoring, in particular to an automatic U-shaped binning algorithm based on quadratic functions and information maximization. Background technique [0002] A credit scoring model is a model that evaluates the credit status of a loan applicant to predict the applicant's probability of serious default or bad debt in the future. Credit scoring models play a very important role in credit risk management because the score is often the key basis for credit approval. [0003] Since the credit business of banks and the Internet finance industry has the characteristics of a large number of transactions and rich data, it is determined that an intelligent and probabilistic management model is required for it. The credit scoring model uses modern mathematical statistical model technology to extract information that can predict the probability of an applicant's fraudulent credit overdue through in-depth mining ...

AI Technical Summary

Problems solved by technology

[0008] The monotonic binning method forcibly reverses some U-shaped characteristic variables into a monotonic trend, which cannot be explained in the actual business sense;

[0009] The monotonic binning method changes the trend of U-shaped structural feature variables, and there will be a large loss in the information value IV (information value) of the binning result, which will affect the final effect of the model

Images