Feature binning algorithm based on decision tree

Abstract

The invention discloses a feature binning algorithm based on a decision tree. The algorithm comprises the steps of modeling a data sample, combining a feature variable and a target variable, setting alimiting condition, binning the decision tree and generating a binning result. The equipment provided by the invention can apply a machine learning decision tree algorithm to generate an optimal binning result under a certain condition, and the finally generated binning result achieves the optimal representation in data significance and eliminates the interference of subjective consciousness of modeling personnel. 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 personal credit risk assessment in financial scenarios, and in particular relates to a feature binning algorithm based on a decision tree. Background technique [0002] The credit scoring model is based on various historical credit data of banks or Internet financial customers to obtain credit scores of different grades. According to the customer's credit score, the credit institution can determine whether to grant credit and credit by analyzing the possibility of customer repayment after the loan amounts and interest rates. [0003] Traditionally, banks or financial institutions adopt manual approval methods, and make subjective approval judgments based on the personal experience of the approvers, making the approval decision easily influenced by subjective factors, resulting in inconsistent approval results, unable to quantify the risk level, and unable to achieve risk control. Hierarchical management,...

AI Technical Summary

Problems solved by technology

[0007] 1. The traditional method does not achieve the optimal effect of binning under certain constraints, 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;

[0008] 2. The traditional method is relatively subjective and does not have uniform applicability. The effects of different modeling engineers are not stable.

Images