Random forest model training method and model training control system

Abstract

The invention relates to a random forest model training method and a model training control system. The random forest model training method includes: the model training control system analyzes whether the model training conditions have been met; if the model training conditions have been met, determine Whether it is necessary to perform reconstructive training on the random forest model; if it is necessary to perform reconstructive training on the random forest model, use sample data to perform reconstructive training on the random forest model; if it is not necessary to perform reconstructive training on the random forest model If the model is reconstructed, then the random forest model is corrected using the sample data. The invention can reduce the training times of the random forest model, lighten the system burden and improve the system efficiency.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a random forest model training method and a model training control system. Background technique [0002] In machine learning, random forest is a classifier that uses multiple trees to train and predict sample data. It is a classifier that includes multiple decision trees. Decision tree is the process of classifying data through a series of rules. At present, more and more companies that provide online services (for example, remote insurance, remote claim settlement, online financial management, etc.) use random forests to classify and identify users in their business systems, and then make accurate business recommendations for users based on the identification results. and handle. [0003] However, when there is new data available as sample data for iterative training to improve the accuracy of model recognition, the existing technical solution is to use both old sample...

AI Technical Summary

Problems solved by technology

This kind of training scheme usually performs a reconstructive training once there is new sample data, and the number of training times is large, especially when the data of online business changes frequently, the training is too frequent and the system is overloaded, which affects the online business system Efficiency and effective development of online business

Images