Data analysis method and device based on AB test result, equipment and medium

Abstract

The invention discloses a data analysis method and device based on AB test results, equipment and a medium, and relates to the technical field of computers, in particular to the field of machine learning. According to the specific implementation scheme, the method comprises the steps: in a first time period, providing a test application and a current application for an experimental group and a control group respectively for AB testing, wherein at least one difference application function exists between the test application and the current application; according to the retention rates of the experimental group and the control group in the first time period, predicting the retention rate index values of the experimental group and the control group in the second time period, wherein the first time period is smaller than the second time period; calculating a retention rate index absolute difference value between the experimental group and the control group according to the retention rate index values of the experimental group and the control group in the second time period; according to the absolute difference value of the retention rate index, determining the effect of the difference application function in the test application, so the effect of the test application with the updated application function can be comprehensively measured for a long time.

Description

technical field [0001] The present disclosure relates to the field of computer technology, in particular to the field of machine learning, and in particular to a data analysis method, device, device and medium based on AB test results. Background technique [0002] With the rapid development of computer technology, it is necessary to continuously update the application functions after the application products are launched to meet the increasingly diverse needs of people. Usually, when an application function is updated, an application product effect evaluation is required to analyze in advance the impact of the new application function on the application product. [0003] However, an application product usually includes multiple application functions. For a comprehensive application product, changing a certain application function will increase or decrease the benefits of other application functions. In the prior art, application product effects are usually evaluated based ...

AI Technical Summary

Problems solved by technology

[0003] However, application products usually include multiple application functions. For comprehensive application products, changing a certain application function will increase or decrease the benefits of other application functions.

In the prior art, the effect evaluation of application products is usually performed according to the modified application functions, but it is impossible to conduct comprehensive effect evaluation on application products, which has one-sidedness and limitations

Images