Micro-video popularity prediction method based on attributive classification and multi-angle feature fusion
A technology of attribute classification and feature fusion, applied in character and pattern recognition, instruments, computer components, etc., can solve problems such as inability to meet various needs, inability to fully understand feature asymmetry, etc., and achieve the effect of improving accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0026] In order to achieve a better prediction effect, a comprehensive, automatic and accurate method for predicting the popularity of micro-videos is needed. Research shows that micro-videos with similar features have similar popularity. The embodiment of the present invention proposes a micro-video popularity prediction method based on attribute classification and multi-view feature fusion, see figure 1 , see the description below:
[0027] 101: Using social attribute features to classify micro-videos, assigning micro-videos to different popularity levels, and obtaining the primary popularity range of micro-videos;
[0028] Among them, the classification of popularity levels is related to the popularity scores of the micro-videos in the test set. Arrange the scores of the micro-videos in the training set from high to low, and then distribute the micro-videos to different levels evenly from high to low in popularity scores, and classify this as the popularity level of the t...
Embodiment 2
[0041] The scheme in embodiment 1 is further introduced below in conjunction with specific calculation formulas and examples, see the following description for details:
[0042] 201: Extract 4 kinds of view mode features for a given micro-video, namely: visual features, acoustic features, text features and social attribute features;
[0043] The embodiment of the present invention firstly extracts four common features of micro-video research from a given micro-video, including: visual features, acoustic features, text features and social attribute features.
[0044] 1. Visual features include: color histogram information, object information in the micro-video (can be obtained by convolutional neural network or other methods, and the embodiment of the present invention does not limit this) and aesthetic features.
[0045] 2. Acoustic features include: the music in the micro-video and the features of other main background sounds.
[0046] 3. Text features include: text annotati...
Embodiment 3
[0073] Below in conjunction with concrete experimental data, example, the scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:
[0074] The test data set used in this experiment is a micro video set downloaded from the Vine social networking site (well known to those skilled in the art, the embodiment of the present invention will not repeat this), and the length of the micro video is 6S. The mean square error and Spearman rank correlation value are used to measure the micro-video popularity prediction performance of this method, the mean square error (nMSE) represents the absolute accuracy of prediction, and the Spearman rank correlation value (SRC) represents the ranking accuracy of prediction .
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


