Video feedback method, device, equipment and medium based on query analysis

Abstract

The present invention relates to the field of artificial intelligence, and provides a video feedback method, device, device, and medium based on query analysis, which can obtain a target query with high user attention and a video segment associated with the target query by analyzing query logs, so as to provide The follow-up training model is used to make the trained model more in line with the user's concerns. The trained video content analysis model can analyze and transform the content of the input video segment, and then output the vector representation of the video segment content, which can reflect the video content. The main content of the segment, the vector representation of the target query and the vector representation of the video segment associated with each target query are used to train the target classification model, which can effectively improve the correlation between the question and the feedback video content, and make the classification of the model The effect is better, and then the target video, and then accurate video feedback based on artificial intelligence means. The present invention also relates to block chain technology, and the trained model can be stored on block chain nodes.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a video feedback method, device, equipment and medium based on query analysis. Background technique [0002] The current application field of video content analysis technology is mainly related to entertainment, such as: sports video collection, variety show video editing, etc., and there is still a lack of video editing in the field of professional training. [0003] In addition, when the current training platforms provide video feedback based on the questions raised by users, they are mainly based on the topic of the video, and do not associate the video content with the questions entered by the user, resulting in insufficient detail and accuracy of the video feedback. Low, unable to meet the needs of users. Contents of the invention [0004] The embodiment of the present invention provides a video feedback method, device, device and medium based on q...

AI Technical Summary

Problems solved by technology

[0002] The current application fields of video content analysis technology are mainly related to entertainment, such as: sports video generation highlights, variety show video editing, etc., and video editing in the field of professional training is still relatively lacking

[0003] In addition, when the current training platforms provide video feedback based on the questions raised by users, they are mainly based on the topic of the video, and do not associate the video content with the questions entered by the user, resulting in insufficient detail and accuracy of the video feedback. Low, unable to match the needs of users

Images