Training method and device of service prediction system

Abstract

The embodiment of the invention provides a training method of a service prediction system. In the first stage of the method, rich label information of a source domain is utilized to train a strong feature extractor shared by the source domain and a target domain, and then the strong feature extractor is applied to the target domain; and in the second stage, the strong features extracted from the target domain sample by the trained feature extractor, and the original features and the service labels in the target domain sample are utilized to carry out supervised training on the service prediction model for the target domain object. Therefore, the strong feature extractor trained in the first stage and the business prediction model trained in the second stage form a business prediction system applied to the target domain.

Description

technical field [0001] One or more embodiments of this specification relate to the field of machine learning, and in particular to a training method and device for a business forecasting system. Background technique [0002] With the rise of machine learning, more and more business platforms analyze and evaluate the business objects of their platforms by training machine learning models. For example, e-commerce platforms, social platforms, etc., through training risk assessment models, conduct risk assessments on operational events in the platforms, and identify high-risk operational behaviors that may threaten network security or user information security, such as account theft and traffic attacks , fraudulent transactions, etc., so as to prevent and control them in a timely manner. [0003] Usually, the training of the model relies on a large amount of labeled data. However, in some areas where the labeled data is scarce, it is difficult to train and learn the model. For...

AI Technical Summary

Problems solved by technology

Models trained directly on such data often have problems such as poor performance

Images