Content Descriptor

• US 20150331876A1
• Filed: 05/22/2015
• Published: 11/19/2015
• Est. Priority Date: 11/08/2011
• Status: Active Grant

• Current Assignee: Comcast Cable Communications LLC (Comcast Corporation)
• Original Assignee: Comcast Cable Communications LLC (Comcast Corporation)
• Inventors: Ambwani, Geetu, Rubinoff, Robert
• Granted Patent: US 11,151,193 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/5846 using extracted text
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data
• CPC Class Codes: G06F 16/7867 using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06V 20/47 Detecting features for summarising video content
• CPC Class Codes: G06V 30/10 Character recognition

Abstract

An apparatus, method, system and computer-readable medium are provided for generating one or more descriptors that may potentially be associated with content, such as video or a segment of video. In some embodiments, a teaser for the content may be identified based on contextual similarity between words and/or phrases in the segment and one or more other segments, such as a previous segment. In some embodiments, an optical character recognition (OCR) technique may be applied to the content, such as banners or graphics associated with the content in order to generate or identify OCR'"'"'d text or characters. The text/characters may serve as a candidate descriptor(s). In some embodiments, one or more strings of characters or words may be compared with (pre-assigned) tags associated with the content, and if it is determined that the one or more strings or words match the tags within a threshold, the one or more strings or words may serve as a candidate descriptor(s). One or more candidate descriptor identification techniques may be combined.

Claims

1. (canceled)

2. A method comprising:

receiving, by a computing device, descriptive data for at least a first segment of a content item and a second segment of the content item;

comparing, by the computing device and based at least in part on the descriptive data, first segment words with second segment words, wherein the first segment words comprise a plurality of words from the first segment from the first segment and second segment words comprise one or more words from the second segment;

determining, by the computing device and based at least in part on the comparing, that the first segment words are descriptive of the second segment; and

determining, by the computing device and responsive to determining that the first segment words are descriptive of the second segment, the first segment words as a candidate descriptor of the second segment.

3. The method of claim 2, wherein the second segment precedes the first segment in the content item.

4. The method of claim 2, wherein the determining that the first segment words are descriptive of the second segment comprises determining that the first segment words are descriptive of the second segment based at least in part on salient tag detection and a threshold.

5. The method of claim 2, wherein the determining that the first segment words are descriptive of the second segment comprises determining that the first segment words are descriptive of the second segment based at least in part on the first segment words being within a threshold distance of a commercial break.

6. The method of claim 2, further comprising:

iteratively comparing the second segment words with a plurality of other words from the first segment;

determining, for each of the other words from the first segment, whether one or more of the other words are descriptive of the second segment; and

based on the one or more of the other words being descriptive of the second segment, determining the one or more of the other words as an another candidate descriptor of the second segment.

7. The method of claim 2, wherein the descriptive data comprises a transcript that is associated with at least the first segment, the second segment, and a third segment of the content item, the method further comprising:

comparing third segment words with the second segment words, wherein the third segment words comprise one or more words included in the third segment;

determining that the third segment words are descriptive of the second segment; and

responsive to determining that the third segment words are descriptive of the second segment, determining the third segment words as a second candidate descriptor of the second segment.

8. The method of claim 2, wherein the determining that the first segment words are descriptive of the second segment comprises determining that the first segment words are descriptive of the second segment based on a normalization of a result generated by the comparing.

9. The method of claim 2, wherein the first segment words comprise at least one of a phrase and a sentence, and wherein the determining that the first segment words are descriptive of the second segment comprises determining that the first segment words are descriptive of the second segment based at least in part on the first segment being longer than a threshold.

10. The method of claim 2, further comprising

determining a relationship strength between the descriptive data for the first segment and descriptive data for at least one additional segment within a threshold proximity to the first segment;

and

determining the first segment words based on the relationship strength of the first segment words being below a maximum threshold.

11. A method comprising:

receiving, by a computing device, descriptive data for a segment of a content item;

receiving, by the computing device, at least one tag assigned to the content item;

iteratively comparing, by the computing device, the descriptive data and the at least one tag; and

determining, by the computing device, whether to add the descriptive data to a candidate descriptor for the content item based on the comparing.

12. The method of claim 11, wherein the descriptive data comprises a transcript.

13. The method of claim 11, wherein the comparing the descriptive data and the at least one tag further comprises determining a relationship strength of the descriptive data relative to the at least one tag.

14. The method of claim 13, wherein the determining whether to add the descriptive data to the candidate descriptor for the content item based on the comparing comprises determining to add at least one portion of the descriptive data based on the relationship strength exceeding a minimum strength level.

15. The method of claim 11, further comprising:

determining a relationship strength for the descriptive data relative to descriptive data for at least one additional segment, wherein the additional segment is within a threshold proximity to the segment; and

wherein the determining whether to add the descriptive data to the candidate descriptor for the content item based on the comparing comprises determining whether to add the descriptive data to the candidate descriptor for the content item based on the comparing and the relationship strength of the descriptive data being below a maximum threshold.

16. The method of claim 11, wherein the comparing the descriptive data and the at least one tag comprises comparing the descriptive data and the at least one tag using salient tag detection.

17. The method of claim 16, wherein the salient tag detection is adjusted based on results from a user survey.

18. A method comprising:

determining, by a computing device, a first relationship strength between a first segment of a content item and a second segment of the content item by iteratively comparing words from the first segment with words from the second segment; and

determining, by the computing device, whether to use the words from the first segment as a candidate descriptor for the second segment responsive to the first relationship strength exceeding a minimum threshold using the words from the first segment and the words from the second segment.

19. The method of claim 18, further comprising:

determining a second relationship strength by iteratively comparing words from a third segment with words from the second segment; and

determining whether to use the words from the third segment as the candidate descriptor for the second segment responsive to the second relationship strength exceeding the first relationship strength.

20. The method of claim 18, wherein the determining the first relationship strength comprises determining the first relationship strength by iteratively comparing words from the first segment with words from the second segment using salient tag detection.

21. The method of claim 20, wherein the salient tag detection is adjusted based on results from a user survey.