Video content analysis for automatic demographics recognition of users and videos

Abstract

A video demographics analysis system selects a training set of videos to use to correlate viewer demographics and video content data. The video demographics analysis system extracts demographic data from viewer profiles related to videos in the training set and creates a set of demographic distributions, and also extracts video data from videos in the training set. The video demographics analysis system correlates the viewer demographics with the video data of videos viewed by that viewer. Using the prediction model produced by the machine learning process, a new video about which there is no a priori knowledge can be associated with a predicted demographic distribution specifying probabilities of the video appealing to different types of people within a given demographic category, such as people of different ages within an age demographic category.

Description

CROSS REFERENCE TO RELATED APPLICATIONS[0001]The application claims the benefit of Provisional Application No. 61 / 147,736, filed on Jan. 27, 2009, which is hereby incorporated herein by reference.BACKGROUND[0002]1. Field of Art[0003]The present invention generally relates to the field of digital video, and more specifically, to methods of correlating demographic data with characteristics of video content.[0004]2. Background of the Invention[0005]Video hosting sites, such as YouTube or Google Video, currently have millions of users and tens of millions of videos. Users may sometimes have difficulty in determining which videos would be of interest to them, and may be daunted by the sheer volume of videos available for viewing. Thus, the ability to suggest which videos would be of interest to a given user is highly valuable.[0006]However, conventional systems typically merely rely on external metadata associated with the video, such as keywords or textual video descriptions, to predict...

AI Technical Summary

Problems solved by technology

Users may sometimes have difficulty in determining which videos would be of interest to them, and may be daunted by the sheer volume of videos available for viewing.

However, if the video is new and has not yet been viewed and rated, and if the associated title is “spam” that misrepresents the true content of the video, then the conventional approach produces spurious predictions.

Thus, one shortcoming of conventional approaches is that they rely on external metadata that may be false when assessing the pertinence of a given video to a particular viewer, rather than examining the actual video content itself.

Images