Object data processing method, apparatus, computer equipment and storage medium

Abstract

The present application relates to an object data processing method, device, computer equipment and storage medium. The method involves a decision tree model in the field of machine learning, and the method includes: according to the proportion of the first hit category in the object behavior data corresponding to each portrait dimension of the object portrait data, which belongs to the first hit category implementing the target strategy for the sample object, and the proportion of each portrait dimension The corresponding object behavior data is converted into the strategy effect of the corresponding target strategy; the strategy effect corresponding to each portrait dimension is input into the decision tree model, and the strategy effect corresponding to the combination of portrait dimensions represented by the nodes in the decision tree model is optimized as the goal. The decision tree model is trained to obtain a group division decision tree; based on the strategy effects corresponding to each subdivision group represented by the root node to the leaf node of the group division decision tree, determine the better subdivision group corresponding to the target strategy. Adopting this method can improve the efficiency of mining subdivided group strategy results from test results.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to an object data processing method, device, computer equipment and storage medium. Background technique [0002] A / B testing is to divide the tested objects into Group A and Group B, implement the target strategy for the objects in Group A, and not implement the target strategy for the objects in Group B, and analyze the object behavior data of the tested objects in Group A and Group B difference to determine the most effective strategy for optimizing the metric. [0003] With the development of computer technology and artificial intelligence technology, A / B testing has been widely used. For example, in the Internet product optimization scenario, the test results of the A / B test are used to determine whether to launch the optimization strategy. In the initial stage of Internet product launch, the optimization strategy can often be recognized by the overall tested obje...

AI Technical Summary

Problems solved by technology

Portrait dimensions include city, age, and gender, for example. City attribute values ​​include Beijing, Shanghai, Shenzhen, Guangzhou, etc., age attribute values ​​include 1-100 years old, and gender attribute values ​​include male and female. Due to the large number of image dimensions, Due to the large number of attribute values ​​in the portrait dimension and other factors, the combination of different attribute values ​​often encounters the problem of combination explosion, resulting in computing performance bottlenecks, which in turn leads to low efficiency in mining subdivided group test results from test results

Images