A short video recommendation method based on video content understanding and user dynamic interest

Abstract

The invention discloses a short video recommendation method based on video content understanding and user dynamic interest, and the method comprises the steps of firstly, extracting deep visual features of a video through a deep learning technology, extracting an audio file from the video, and extracting auditory features; fusing the video features, the social features and the user features by using PCA dimension reduction, data standardization and other technologies to obtain deep fusion features for feature representation of historical behaviors of the user; secondly, extracting the influence of historical behaviors on the current interest by using a self-attention mechanism, and learning an interest evolution path on the candidate video by using a recurrent neural network to obtain accurate dynamic interest of the user; and finally, utilizing a multilayer perceptron to carry out click probability prediction and recommendation on the video candidate set. The method is applied to thepersonalized recommendation of the short video, and the recommendation accuracy can be effectively improved by adopting the technical scheme of the invention.

Description

technical field [0001] The invention belongs to the technical field of network video, and in particular relates to a short video recommendation method based on video content understanding and user dynamic interest. Background technique [0002] With the popularity of mobile terminals and the speed up of the Internet, short, flat and fast videos are favored by major platforms and users, and short video platforms are gradually rising, followed by problems of information overload and personalized needs. Massive video is a huge challenge for video consumers and video producers. For video consumers, it is difficult to find the videos that users are really interested in from a large number of videos; for video providers, it is difficult to distribute videos to appropriate users. Due to these urgent needs, the personalized recommendation of mobile short videos has become a hot research topic. [0003] The methods applied to personalized recommendation include content-based recomm...

AI Technical Summary

Problems solved by technology

[0004] Currently, methods for video recommendation only consider the user's preference for video history behavior, but do not explore video content

Compared with the personalized recommendation of ordinary resources, the personalized recommendation of mobile short video has the problem that unstructured video information is difficult to use.

[0005] In addition, there are the following problems: (1) Mobile short videos often do not have titles, descriptions and other information that match the video content; (2) The classification of mobile short videos is only rough, and it is difficult to accurately express user interests; (3) Users Feedback data is sparse

Images