Method and device for acquiring user information

Abstract

The embodiment of the invention provides a method and a device for acquiring user information and relates to the technical field of a computer. The invention aims to increase the stability of a forecasting model for calculating the user characteristic information by regularly updating a training sample of the forecasting model. According to the technical scheme adopted by the embodiment of the invention, the method comprises the following steps: collecting a multimedia sample set which comprises a multimedia sample for distinguishing the user characteristic information; counting and recording the viewing users of the multimedia sample set and selecting the training user sample which is the user sample with clear user characteristic information; training a user information acquiring model by utilizing the training user sample; and acquiring the characteristic information of the target user by adopting the user information acquiring model. The method provided by the invention is mainly used for acquiring the user characteristic information.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a method and device for acquiring user information. Background technique [0002] With the advent of the cloud era, big data (Big data) has also attracted more and more attention. A lot of intelligent, in-depth and valuable information can be obtained through the analysis of big data. At present, in the field of video, major video websites are paying more and more attention to users' browsing experience on the website, such as whether the content of the video is rich, whether users can find the video resources they need more easily, etc. For this reason, video websites Targeted video content recommendations will be made for different users to make it easier for users to find the video content they want to watch. To achieve targeted push information to users, it is first necessary to determine the characteristic information of different users, suc...

AI Technical Summary

Problems solved by technology

However, during the process of proposing the present invention, the applicant found that the currently used calculation models for predicting user characteristic information are mostly one-time models, which are used after training. It is more suitable for large cases, but in the video field, although the attributes of a single movie will not change, the popular videos in each time period will change, and the user base in each time period will also continue to change

These changing factors may affect the final calculation results of the prediction model. It can be seen that the existing prediction models have great instability when calculating the feature information of movie viewers.

Images