Content contribution degree evaluation method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a content contribution degree evaluation method and device, electronic equipment and a readable storage medium. The method comprises the steps that a content feature set generated by a user on a target application in a target statistical period is obtained; determining all feature subsets corresponding to the content feature set, and sequentially inputting each feature subset into a pre-trained retention prediction model for prediction to obtain a user retention result predicted based on each feature subset; and according to the retention result of the user predicted based on each feature subset, the content feature set and a preset contribution degree analysis model, obtaining the retention behavior contribution degree of each content in the content feature set to the user. According to the embodiment of the invention, the influence degree of different content features on the retention behavior of the user can be reversely determined in combination with the predicted retention condition, an operation strategy and cost release can be specifically formulated based on the contribution degree, and the use viscosity of the user to the target application can be further improved.

Description

technical field [0001] The present invention relates to the field of live broadcast technology, in particular, to a content contribution assessment method, device, electronic equipment and readable storage medium. Background technique [0002] In the field of Internet technology, various Internet products have emerged as the times require. By judging whether users are retained or not, the degree of stickiness of various Internet products to users can be determined, which can assist product development and operation and optimize operating costs; [0003] At present, the existing methods for user retention are mainly based on statistical methods. On the one hand, this method needs to determine the weight of different observation indicators first, and this weight will change with the actual behavior of users, so that the retention behavior analysis results On the other hand, since users of Internet products interact with a variety of different content, it is difficult to accura...

AI Technical Summary

Problems solved by technology

[0003] At present, the existing methods for user retention are mainly based on statistical methods. On the one hand, this method needs to determine the weight of different observation indicators first, and this weight will change with the actual behavior of users, so that the retention behavior analysis results On the other hand, since users of Internet products interact with a variety of different content, it is difficult to accurately split out content that cannot be accurately analyzed if only indicators such as viewing time are used for statistical analysis. Contribution to user retention behavior, it is easy to draw wrong conclusions

Images