Multimedia classification method and device and computer readable storage medium

Abstract

The embodiment of the invention discloses a multimedia classification method and device and a computer readable storage medium. According to the embodiment of the invention, a multimedia sample and ainitial label set corresponding to the multimedia sample can be obtained, frequent labels in the initial label set are selected according to the occurrence frequency of each label in the initial labelset to obtain the frequent label set, and the semantic similarity between the labels in the frequent label set is calculated. A target label set corresponding to the multimedia sample is determined from the frequent label set according to a calculation result, prediction accuracy and prediction recall rate of each label in the target label set is calculated and the target label set is filtered according to the calculation result to obtain a multimedia classification data set; To-be-classified multimedia is classified based on the multimedia classification data set to obtain a classification label of the to-be-classified multimedia. According to the scheme, the multimedia classification efficiency can be effectively improved.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to a multimedia classification method, device and computer-readable storage medium. Background technique [0002] With the rapid development of the Internet, multimedia data on the Internet, including video, audio, text, etc., has become one of the hot research issues due to the continuous growth of the number. A large amount of multimedia information causes data accumulation and cannot be processed in time. Therefore, quickly and effectively browsing and classifying a large amount of multimedia data is very important for improving user experience and discovering potential commercial value. [0003] In common video data, users often describe and comment on video information and provide personalized label data, etc., thus forming rich text information related to online video. The text information needs to be processed to obtain a video classification data set before ...

AI Technical Summary

Problems solved by technology

The text information needs to be processed to obtain a video classification data set before it can be used for video classification. However, the existing method of video classification data set is to manually select the appropriate label from a bunch of classification labels, which consumes a lot of time. Manpower and material resources, and it is easy to misjudge the difficulty of label prediction

Images