Business data processing method and computer equipment

Abstract

The business data processing method and computer device provided by the embodiments of the present invention can effectively improve the original model in the step-by-step iteration by using the design of the algorithm in the scenario where the modeling data can be effectively obtained in both the first business area and the second business area. There are weights of the samples in the model that are similar to those in the second business area, while the weights of the non-similar samples in the original model are reduced. In the iteration, through the set iteration early stop mechanism, the loop is terminated without a significant improvement in model performance in multiple rounds of iterations. In this way, on the one hand, the subjectivity of manually setting the number of iterations is avoided, and on the other hand, when the loop is terminated, the model with the best performance in multiple rounds of iterations can be selected for actual deployment, which solves the need for deployment iterations. All models in to output the complexity of the model results. In this way, the generalization performance of the risk control model can be improved, thereby ensuring that the prediction accuracy will not be biased.

Description

technical field [0001] The present invention relates to the technical field of data feature processing, in particular to a business data processing method and computer equipment. Background technique [0002] With the development of global economic integration, the financial and credit business between different regions is also developing interactively. When processing financial credit business, it is necessary to realize big data risk control processing. In order to ensure the compatibility of the big data risk control system in different regions, it is necessary to model the risk control model based on sample data. However, the data scale of sample data in different regions is different, which will lead to different regions The modeling of the risk control model may lead to deviations in the prediction accuracy of the risk control model. Contents of the invention [0003] In order to improve the above problems, the present invention provides a business data processing ...

AI Technical Summary

Problems solved by technology

In order to ensure the compatibility of the big data risk control system in different regions, it is necessary to model the risk control model based on sample data. However, the data scale of sample data in different regions is different, which will lead to different regions Risk control model modeling may lead to deviations in the prediction accuracy of the risk control model

Images