IDENTIFYING REFERENCE CONTENT THAT INCLUDES THIRD PARTY CONTENT

摘要

A reference verification module identifies content items stored in a content repository that include at least a portion of the content included in a reference content item provided by an entity. The reference verification module classifies the reference content item based on the uneven distribution of the percentages of the identified content items that include each of the claimed portions. Specifically, if the average difference between the percentage of content items that include a given portion and the percentages of the remaining content items is above a threshold, then the reference verification module classifies the reference content item as having a higher than threshold likelihood of including third party content.

主权项

1. A computer implemented method for identifying reference video content that includes third party content, the method comprising:
identifying reference video content provided by an entity to a content hosting service; identifying a plurality of content items received from users of the content hosting service that include at least one portion of the reference video content; for each portion of the reference video content, determining a number of the identified plurality of content items that include that portion of the reference content; and classifying the reference video content based on a relative difference between the determined numbers for the portions of the reference video content, the relative difference indicating a likelihood that the reference video content includes third party content.