Method and apparatus for conducting context sensitive search with intelligent user interaction from within a media experience

• US 9,710,545 B2
• Filed: 12/20/2012
• Issued: 07/18/2017
• Est. Priority Date: 12/20/2012
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Wouhaybi, Rita H., Mo, Stanley, Szilagyi, Victor
• Primary Examiner(s): PULLIAS, JESSE SCOTT
• Application Number:
  US13/722,274
• Publication Number:
  US 20140180672A1
• Time in Patent Office: 1,671 Days
• Field of Search: 704 1, 704 9, 704 10, 704257
• CPC Class Codes: G06F 16/3344 using natural language analysis

Abstract

In some embodiments, the invention involves context based search engine using a user selected term within a media experience. A natural language processor module is configured to provide context based keywords related to the search term and from within the media experience. In some embodiments, a proximity based statistical analysis is used to derive the keywords. The keywords are provided to at least one content browser or other search engine(s) to effect the search. In some embodiments, a machine learning module is communicatively coupled to the natural language processer to further refine the context for selecting relevant keywords. The context search engine, natural language processor module, machine learning module and search engine may reside on the same computing device or be distributed among a variety of local, remote and cloud devices for processing. Other embodiments are described and claimed.

Claims

1. A system for providing a contextual based search, comprising:

a computing device configured to provide a media experience to a user, the computing device further comprising;

an output display;

an input device for selecting a search term within the media experience; and

a context search engine communicatively coupled to a natural language processor module and a content browser, wherein the context search engine is configured to provide the search term to the natural language processor, and provide keyword results received from the natural language processor to the content browser;

wherein the keyword results are provided to the user as one or more of videos, images, or graphics based on a first historical preference of the user, and as a textual reference based on a second historical preference of the user, and

wherein topics of interest to the user are to be determined based on historical use of multiple forms of media by the user over a predetermined period of time, and the topics of interest are to be used to prioritize the keyword results.

2. The system recited in claim 1, wherein the natural language processor module resides on at least one of the computing device, a content server, a cloud server, a remote server and a local computing device, wherein when the natural language processor module does not reside on the computing device, the natural language processor may be accessed via at least one of a direct connection, a wired connection and a wireless connection.

3. The system as recited in claim 1, wherein the natural language processor module is configured to use proximity based statistical analysis in identifying the keyword results to be provided to the context search engine.

4. The system as recited in claim 3, wherein the natural language processor module is configured to perform the proximity based statistical analysis on a portion of text within the media experience, wherein the portion of text is within a specified proximity of the search term.

5. The system as recited in claim 1, further comprising a machine learning module communicatively coupled to the natural language processor module.

6. The system recited in claim 5, wherein the natural language processor module is configured to perform a proximity based statistical analysis on a portion of text within the media experience, wherein the portion of text is within a specified proximity of the search term, and wherein the machine learning module is configured to dynamically adjust the size of the data portion based on historical contextual information.

7. The system as recited in claim 1, wherein the computing device is further configured to display results returned from the content browser, wherein the content browser is configured to perform searches using the provided keywords.

8. The system as recited in claim 7, wherein the context search engine is configured to provide the search term to at least one additional content browser and aggregate results from the content browser and the at least one additional content browser and display the results to the user.

9. A system for providing a contextual based search, comprising:

a context search engine communicatively coupled to a natural language processor module and at least one content browser, wherein the context search engine is configured to provide a user selected search term to the natural language processor, and provide keyword results received from the natural language processor to the content browser;

wherein the keyword results are provided to the user as one or more of videos, images, or graphics based on a first historical preference of the user, and as a textual reference based on a second historical preference of the user, and

wherein topics of interest to the user are to be determined based on historical use of multiple forms of media by the user over a predetermined period of time, and the topics of interest are to be used to prioritize the keyword results.

10. The system as recited in claim 9, wherein the selected search term is to be selected from closed captioning text of one of a television program, broadcast program, streaming media and recorded media.

11. The system as recited in claim 10, wherein the closed captioning text is to be selected using one of a remote control device and a secondary computing device.

12. The system as recited in claim 11, wherein the context search engine is configured to provide results from the content browser for display on at least one of the television and the secondary device.

13. The system as recited in claim 10, wherein the natural language processor module is configured to use proximity based statistical analysis in identifying the keyword results to be provided to the context search engine.

14. The system as recited in claim 13, wherein the natural language processor module is configured to perform the proximity based statistical analysis on a portion of text within an available portion of the closed captioning text, wherein the portion of text is within a specified proximity of the search term.

15. The system as recited in claim 9, further comprising a machine learning module communicatively coupled to the natural language processor module, wherein the natural language processor module is configured to perform a proximity based statistical analysis on a portion of text within a specified proximity of the search term, and wherein the machine learning module is configured to dynamically adjust the size of the text portion based on historical contextual information.

16. At least one non-transitory computer readable storage medium having embodied thereon instructions which when executed on a machine cause the machine to:

responsive to user selection of a search term within a media experience, send the search term by a context search engine to a natural language processor module;

scan content within proximity of the search term by the natural language processor module, and determine keywords associated with the search term; and

provide the keywords to at least one content browser for search;

wherein keyword results are provided to a user as one or more of videos, images, or graphics based on a first historical preference of the user, and as a textual reference based on a second historical preference of the user, and

wherein topics of interest to the user are to be determined based on historical use of multiple forms of media by the user over a predetermined period of time, and the topics of interest are to be used to prioritize the keyword results.

17. The medium as recited in claim 16, further comprising instructions to:

display results of the search by the at least one content browser to the user.

18. The medium as recited in claim 17, wherein the search term is selected from one of an electronic document accessible to a user device and closed captioning text corresponding to one of a television program, broadcast program, streaming media and recorded media available on a viewing device.

19. The medium as recited in claim 18, wherein results of the search are displayed on one of a user device and a secondary user device.

20. The medium as recited as in claim 16, further comprising instructions to:

identify the keywords to be provided to the context search engine by the natural language processor module, wherein the identifying uses proximity based statistical analysis.

21. The medium as recited in claim 20, further comprising instructions to:

perform the proximity based statistical analysis on a portion of text within the media experience, wherein the portion of text is within a specified proximity of the search term.

22. The medium as recited in claim 16, further comprising instructions to:

perform a proximity based statistical analysis on a portion of text within the media experience, wherein the portion of text is within a specified proximity of the search term; and

dynamically adjust the size of the data portion, by a machine learning module, wherein the adjusting is based on historical contextual information.

23. The medium as recited in claim 22, further comprising instructions to:

provide the search term to at least one additional content browser session and aggregate results from the content browser and the at least one additional content browser session; and

display the aggregated results to the user.