Television key phrase detection

• US 9,876,848 B1
• Filed: 02/21/2014
• Issued: 01/23/2018
• Est. Priority Date: 02/21/2014
• Status: Active Grant

• Current Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Original Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Inventors: Carroll, Andrew John, Whitcomb, Jr., Richard Douglas, Fu, Wangfan, Jianrungsang, Thanabordi
• Primary Examiner(s): Etienne, Ario
• Assistant Examiner(s): Kassa, Elizabeth
• Application Number:
  US14/187,056
• Time in Patent Office: 1,432 Days
• Field of Search: 709203
• 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 2218/12 Classification; Matching
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

Images of key phrases or hashtags appear on televised feeds. Image processing techniques, such as feature locating algorithms or character recognition algorithms, can be used to locate the images of key phrases in the images. Then, character recognition algorithms can be used to generate a list of candidate key phrases for the key phrase in image format. However, identification of the key phrase in image format is not completely accurate with conventional methods. Social media content items associated with the televised feed are used to filter the list of candidate key phrases. Using known information about the televised feed as well as about key phrases in text format in the social media content items, candidate key phrases in the list of candidate key phrases can be scored and, thus, a final candidate key phrase selected based on the scores.

Claims

1. A computer-executed method comprising:

accessing an image comprising an unidentified key phrase in image format;

processing a portion of the image associated with the unidentified key phrase using a character recognition algorithm to generate a list of candidate key phrases in text format;

accessing, by a computer processor, a plurality of social media content items in text format authored by users of a social networking system, the social media content items each comprising at least one key phrase previously associated with the image and at least one of the candidate key phrases;

determining, by the computer processor, for each candidate key phrase, a count of the number of the content items comprising that candidate key phrase and at least one of the key phrases previously associated with the image;

selecting, by the computer processor, one of the candidate key phrases as being the unidentified key phrase based on the determined count of the content items in each set; and

sending content associated with the image to one or more users of the social networking system based on the social media content items including the selected candidate key phrase.

2. The computer-executed method of claim 1, wherein accessing the image comprises extracting the image from a segment of time based media comprising a television show.

3. The computer-executed method of claim 1, wherein the unidentified key phrase comprises a formatting element such as a hash character.

4. The computer-executed method of claim 1, wherein the candidate key phrases in text format are filtered based on an identified language of the text in the social media content items.

5. The computer-executed method of claim 1, further comprising generating additional candidate key phrases using social media content items associated with authors of the accessed social media content items.

6. The computer-executed method of claim 1, wherein the social media content items comprise a plurality of unique second key phrases.

7. The computer-executed method of claim 1, wherein the image is associated with an airing time stamp, and wherein each social media content item in the plurality of social media content items comprises a broadcast time stamp within a threshold amount of time from the airing time stamp.

8. The computer-executed method of claim 1, wherein the second key phrase is not previously known to be associated with the image.

9. A non-transitory computer readable storage medium product for identifying an unidentified key phrase in an image comprising computer program instructions that when executed by a processor cause the processor to:

access an image comprising an unidentified key phrase in image format;

process a portion of the image associated with the unidentified key phrase using a character recognition algorithm to generate a list of candidate key phrases in text format;

access, by a computer processor, a plurality of social media content items in text format authored by users of a social networking system, the social media content items each comprising at least one phrase previously associated with the image and at least one of the candidate key phrases;

determine, by the computer processor, for each candidate key phrase, a count of the number of the content items comprising that candidate key phrase and at least one of the key phrases previously associated with the image;

select one of the candidate key phrases as being the unidentified key phrase based on the determined count of the content items in each set; and

send content associated with the image to one or more users of the social networking system based on the social media content items including the selected candidate key phrase.

10. The computer readable storage medium product of claim 9, wherein the computer program code that when executed by the processor further causes the processor to access the image extracting the image from a segment of time based media comprising a television show.

11. The computer readable storage medium product of 9, wherein the unidentified key phrase comprises a formatting element such as a hash character.

12. The computer readable storage medium product of 9, wherein the candidate key phrases in text format are filtered based on an identified language of the text in the social media content items.

13. The computer readable storage medium product of 9, wherein the computer program code that when executed by the processor further causes the processor to generate additional candidate key phrases using social media content items associated with authors of the accessed social media content items.

14. The computer readable storage medium product of 9, wherein the social media content items comprise a plurality of unique second key phrases.

15. The computer readable storage medium product of 9, wherein the image is associated with an airing time stamp, and wherein each social media content item in the plurality of social media content items comprises a broadcast time stamp within a threshold amount of time from the airing time stamp.

16. The computer readable storage medium product of 9, wherein the second key phrase is not previously known to be associated with the image.

17. A system comprising a processor and a memory storing computer program instructions for identifying an unidentified key phrase in an image that when executed by the processor cause the processor to:

access an image comprising an unidentified key phrase in image format;

process a portion of the image associated with the unidentified key phrase using a character recognition algorithm to generate a list of candidate key phrases in text format;

access, by a computer processor, a plurality of social media content items in text format authored by users of a social networking system, the social media content items each comprising at least one key phrase previously associated with the image and at least one of the candidate key phrases;

determine, by the computer processor, for each candidate key phrase, a count of the number of the content items comprising that candidate key phrase and at least one of the key phrases previously associated with the image;

select one of the candidate key phrases as being the unidentified key phrase based on the determined count of the content items in each set; and

send content associated with the image to one or more users of the social networking system based on the social media content items including the selected candidate key phrase.

24. The system of claim 17, wherein the second key phrase is not previously known to be associated with the image.

18. The system of 17, wherein the computer program instructions that when executed by the processor further causes the processor to access the image extracting the image from a segment of time based media comprising a television show.

19. The system of 17, wherein the unidentified key phrase comprises a formatting element such as a hash character.

20. The system of 17, wherein the candidate key phrases in text format are filtered based on an identified language of the text in the social media content items.

21. The system of 17, wherein the computer program code that when executed by the processor further causes the processor to generate additional candidate key phrases using social media content items associated with authors of the accessed social media content items.

22. The system of 17, wherein the social media content items comprise a plurality of unique second key phrases.

23. The system of 17, wherein the image is associated with an airing time stamp, and wherein each social media content item in the plurality of social media content items comprises a broadcast time stamp within a threshold amount of time from the airing time stamp.