Media content matching using contextual information

A system comprising:

• a computer processor;
• and
• a non-transitory computer-readable storage medium storing computer executable components configured to cause the computer processor to perform:
• determining whether a similarity value between characteristics of a portion of a probe media content item and characteristics of a portion of a reference media content of a reference media content item exceeds a threshold, wherein the threshold depends on the characteristics of the portion of the probe media content item and the characteristics of the portion of the reference media content;
• in response to determining that the similarity value exceeds the threshold, determining that the probe media content item includes the portion of reference media content of the reference media content item;
• upon determining that the probe media content item includes the portion of the reference media content, receiving a pair of media content items comprising the probe media content item and the reference media content item;
• receiving metadata associated with the pair, the metadata associated with the pair providing information about the probe media content item, the reference media content item, and the portion of reference media content reused in the probe media content;
• classifying the pair into a reuse group of a plurality of reuse groups based on the metadata associated with the pair, each of the plurality of reuse groups associated with a corresponding amount of reuse different from other reuse groups, the corresponding amount of reuse used for determining whether or not the probe media content item is to be flagged for removal;
• comparing first amount of the portion of reference media content to a second amount of reuse associated with the reuse group into which the pair is classified;
• and
• responsive to the first amount of the portion of reference media content being greater than the second amount of reuse associated with the reuse group, flagging the probe media content item for removal.

A non-transitory computer-readable medium having instructions stored thereon that, in response to execution, cause a system including a processor to perform steps comprising:

• determining whether a similarity value between characteristics of a portion of a probe media content item and characteristics of a portion of a reference media content of a reference media content item exceeds a threshold wherein the threshold depends on the characteristics of the portion of the probe media content item and the characteristics of the portion of the reference media content;
• in response to determining that the similarity value exceeds the threshold, determining that the probe media content item includes the portion of reference media content of the reference media content item;
• upon determining that the probe media content item includes the portion of the reference media content, receiving a pair of media content items comprising the probe media content item and the reference media content item;
• receiving metadata associated with the pair, the metadata associated with the pair providing information about the probe media content item, the reference media content item, and the portion of reference media content reused in the probe media content;
• classifying the pair into a reuse group of a plurality of reuse groups based on the metadata associated with the pair, each of the plurality of reuse groups associated with a corresponding amount of reuse different from other reuse groups, the corresponding amount of reuse used for determining whether or not the probe media content item is to be flagged for removal;
• comparing first amount of the portion of reference media content to a second amount of reuse associated with the reuse group into which the pair is classified;
• and
• responsive to the first amount of the portion of reference media content greater than the second amount of reuse associated with the reuse group, flagging the probe media content item for removal.