Video recommendation method and system

Abstract

The invention relates to a video recommendation method and a system. The video recommendation method comprises steps of: according to a video recommendation log, counting the exposure number and the clicking number of each of the first type of videos and the exposure number of each of the second type of videos; using a real clicking rate of each of the first type of videos as a comprehensive clicking rate thereof, using a potential clicking rate of each of the second type of videos as a comprehensive clicking rate thereof and obtaining a video recommendation sequence according to the size sequence of the comprehensive clicking rates; according to an accumulated exposure proportion of the nth video in the video recommendation sequence, carrying out traversal counting on videos in the video recommendation sequence, and using an accumulated exposure proportion which is not more than a preset high-quality flow proportion and has the biggest value as the optimal accumulated exposure proportion; using a video with the optimal accumulated exposure proportion and all videos in front of the video in the video recommendation sequence as the optimal video sequence; and judging whether the optimal accumulated exposure proportion is equal to the high-quality flow proportion and using the optimal video sequence as a high-quality flow video group for recommendation. According to the invention, the clicking rate of the video can be improved.

Description

technical field [0001] The present invention relates to the technical field of video recommendation, in particular to a video recommendation method and system. Background technique [0002] The vigorous development of the Internet industry has spawned a large number of emerging industries, and online video playback is one of them. The working process of the current common network video playback technology is as follows: first, video exposure, that is, video websites or Internet TV stations and other playback terminals display several video links on the user interface for users to choose; It is necessary to click a certain video link on the user interface; the next step is to play the video, that is, the player sends the video information corresponding to the video link clicked by the user to the user interface in sequence through the network for playback. The video here can be a film and television program or an advertisement video. [0003] As a business model, online vid...

AI Technical Summary

Problems solved by technology

It can be seen that all or more of the existing technologies consider the profit of the player, but ignore the needs of users, which is likely to cause the problem of shrinking user base

The technical reason for this problem is that the principle of bidding screening cannot satisfy users' needs to the maximum extent, resulting in a gradual decrease in the click-through rate of users on videos

Images