Target object evaluation method based on multi-score card fusion and related equipment thereof

Abstract

The invention belongs to the field of artificial intelligence, and relates to a target object evaluation method based on multi-score card fusion and related equipment thereof, and the method comprises the steps: carrying out the data cleaning according to the distribution state of feature variables contained in the historical data of a target object, and then carrying out the feature variable screening to obtain an original feature variable set; performing clustering operation on the original feature variable set to obtain a plurality of mutually exclusive target feature variable sets; constructing a plurality of target LR models based on each target feature variable set and generating a model precision value; and generating a plurality of score card models based on each target LR model, performing score card fusion according to each score card model and the corresponding model precision value to obtain a target score card model, and outputting a target score value based on the target score card model. In addition, the invention also relates to a block chain technology, and the data determined as the private information can be stored in a block chain. According to the method, the model risk caused by characteristic variable offset can be reduced, and the model precision is ensured while the model stability is improved.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, and specifically relates to a target object evaluation method, device, computer equipment and storage medium based on the fusion of multiple scorecards. Background technique [0002] Scorecard models are widely used in risk assessment and control in the financial industry, such as pre-loan application scorecards, loan behavior scorecards, post-loan collection scorecard models, etc. During the training process of the scorecard model, it is necessary to obtain the characteristic variables as the model input. On the one hand, when obtaining the characteristic variables, the characteristic variables need to be screened to reduce the information redundancy of the characteristic variables. On the other hand, the model fitting process is highly correlated. Feature variables will be partially eliminated, which will lose the diversity of features to a certain extent, thereby redu...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiment of the present application is to propose a target object evaluation method, device, computer equipment, and storage medium based on multi-scorecard fusion, so as to solve the problem of the accuracy of the scorecard model caused by the loss of feature diversity due to the elimination of feature variables in the prior art. decrease, and the problem of reduced stability of the scorecard model due to the long time span of the target variable

Images