Intelligent video playing method and device based on big data and electronic equipment

Abstract

The invention provides an intelligent video playing method based on a big data model; the method comprises the steps: responding to a current user request, searching target video content, determining an associated user, obtaining the data of a sample user, supervising, learning and training a frame screening model, collecting the playing result data of the associated user for the target video content, and generating a frame playing curve; and selecting a target frame picture by utilizing the attribute information of the current user, the operation duration for triggering the video content request and the playing amount of each frame picture, playing a frame picture after the target frame picture; and displaying a frame picture playing curve and an operation component, so that the user can select the frame picture to be played by adjusting the operation component. According to the method, the target frame picture is selected for the current user through the playing result data of the big data collection video content, and the target frame picture is directly used as the picture after the first frame is played, so that the user can quickly browse the required picture, the condition that a large amount of user time is occupied when a non-key frame is played is avoided, and the pushing accuracy is improved.

Description

technical field [0001] The present application relates to the Internet field, and in particular to a method, device and electronic equipment for intelligent video playback based on big data. Background technique [0002] With the development of big data and artificial intelligence, the level of user data mining is getting higher and higher, and video recommendation and playback has become a hot topic. At present, most of the video recommendations have adopted artificial intelligence algorithms, and push relevant content to users according to their preferences. of video content. [0003] However, after the user browses the pushed video, the user clicks to start playing, and then browses the video content. However, although this push method can achieve accurate push to a certain extent, there is still room for improvement in its accuracy. [0004] It is necessary to provide a new video playback method to improve the accuracy of push and improve the effect of push. [0005] A...

AI Technical Summary

Problems solved by technology

[0005] An analysis of the existing technology found that this is mainly because, at present, when recommending videos to users, most users actively click to start, and then play according to the time axis of the video. The images that users want to browse most often can only be seen after playing according to the time axis. Since the user's browsing behavior itself is a kind of resource, the key frame images are hidden in the time axis, resulting in some irrelevant frame images occupying a large amount of users. time, when the user does not see the picture of interest within a certain period of time, they will often quit the video content, which brings a large opportunity cost. This is essentially because, from the dimension of the frame picture, accurate push is not realized

Images