Media search broadening

• US 10,289,737 B1
• Filed: 06/10/2013
• Issued: 05/14/2019
• Est. Priority Date: 04/30/2007
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Sahami, Mehran, Gossweiler, III, Richard C., Patel, Manish G., Blackburn, John, Brown, David A., Gupta, Neha, Taylor, Thomas H.
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Perez-Arroyo, Raquel
• Application Number:
  US13/914,026
• Time in Patent Office: 2,164 Days
• Field of Search: 707708
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A computer-implemented method includes receiving a search request at a central information provider, generating a query containing media-related terms relating to the search request, and submitting the query to a search engine having access to media-related search results, wherein the query is broader than the received search request, and transmitting media-related search results generated in response to the query by the search engine, to a remote device.

Claims

1. A computer-implemented method, comprising:

receiving, at a central information provider, a search request input by a user at a user device;

parsing the search request to identify one or more terms in the search request;

in response to receiving the search request, and without receiving further input from the user about the search request, identifying a plurality of documents that are each determined to include at least a particular term of the one or more terms in the search request;

classifying the search request as a media-related request based at least on the identified plurality of documents, the classifying comprising;

generating one or more context identifiers for the search request based at least on the identified plurality of documents that include the particular term, wherein each context identifier contains words that occur in context with one or more terms in the search request in one or more of the plurality of documents; and

using the one or more context identifiers, identifying that the one or more terms in the search request and the one or more context identifiers are associated with a media-related request;

modifying the search request comprising

     1) adding one or more terms to the request to include a media-related category, and

     2) discarding one or more words from the search request input by the user, wherein adding the one or more terms to the request to include the media-related category biases search results toward returning media-related results;

generating search results for the modified search request;

formatting the search results in a structured display that is pre-assigned to media related requests by generating markup code for display on the user device, wherein the formatting includes a title and an indication relating to how to access media relating to the title; and

providing, to the user at the user device, the formatted search results for display.

2. The computer-implemented method of claim 1, wherein the modified search request includes at least one or more terms that were not in the received search request.

3. The computer-implemented method of claim 1, wherein generating search results for the modified search request comprises, in response to classifying the received search request as a media-related search request, selecting a subset of documents from a corpus of documents identified as being media-related documents.

4. The computer-implemented method of claim 3, further comprising, in response to classifying the received search request as a media-related request, formatting the representations for the identified subset of documents to include an electronic program guide grid that includes an episode of a program that is responsive to the received search request in a cell of the grid.

5. The computer-implemented method of claim 1, wherein generating the one or more context identifiers for the search request comprises supplying one or more words or word groups to a machine learning system that has been trained on media-related content.

6. The computer-implemented method of claim 5, wherein the machine learning system includes a graph relating media-related words or word groups by normalized scores.

7. The computer-implemented method of claim 1, wherein generating the one or more context identifiers for the search request comprises submitting one or more words in the search request to a learning system to generate a context vector indicative of the one or more words'"'"' relationship to media-related words in a graph.

8. The computer-implemented method of claim 1, wherein generating the one or more context identifiers for the search request based at least on the plurality of documents comprises determining, in a graph of word and word group relationships in a machine learning system, a distance between the one or more terms in the search request and the one or more other terms that appear in a corresponding one of the plurality of documents.

9. The computer-implemented method of claim 1, wherein the modified search request includes the search request, a Boolean connector, and a media-related term that defines a type of media.

10. The computer-implemented method of claim 9, wherein the media-related term is selected from the group consisting of television, TV, and movie.

11. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving, at a central information provider, a search request input by a user at a user device;

parsing the search request to identify one or more terms in the search request;

in response to receiving the search request, and without receiving further input from the user about the search request, identifying a plurality of documents that are each determined to include at least a particular term of the one or more terms in the search request;

classifying the search request as a media-related request based at least on the identified plurality of documents, the classifying comprising;

generating one or more context identifiers for the search request based at least on the identified plurality of documents that include the particular term, wherein each context identifier contains words that occur in context with one or more terms in the search request in one or more of the plurality of documents; and

using the one or more context identifiers, identifying that the one or more terms in the search request and the one or more context identifiers are associated with a media-related request;

modifying the search request comprising

     1) adding one or more terms to the request to include a media-related category, and

     2) discarding one or more words from the search request input by the user, wherein adding the one or more terms to the request to include the media-related category biases search results toward returning media-related results;

generating search results for the modified search request;

formatting the search results in a structured display that is pre-assigned to media related requests by generating markup code for display on the user device, wherein the formatting includes a title and an indication relating to how to access media relating to the title; and

providing, to the user at the user device, the formatted search results for display.

13. The system of claim 11, wherein generating the one or more context identifiers for the search request comprises submitting one or more words in the search request to a learning system to generate a context vector indicative of the one or more words'"'"' relationship to media-related words in a graph.

14. The system of claim 11, wherein generating the one or more context identifiers for the search request based at least on the plurality of documents comprises determining, in a graph of word and word group relationships in a machine learning system, a distance between the one or more terms in the search request and the one or more other terms that appear in a corresponding one of the plurality of documents.

15. The system of claim 11, wherein the operations further comprise generating a list of suggested queries that are semantically related to the search request, and providing the list of suggested queries to the user.

12. The system of 11, wherein the modified search request includes at least one or more terms that were not in the received search request.

16. A non-transitory computer-readable medium storing software comprising instructions executable by one or more computers which, upon such execution, cause the one or more computers to perform operations comprising:

receiving, at a central information provider, a search request input by a user at a user device;

parsing the search request to identify one or more terms in the search request;

in response to receiving the search request, and without receiving further input from the user about the search request, identifying a plurality of documents that are each determined to include at least a particular term of the one or more terms in the search request;

classifying the search request as a media-related request based at least on the identified plurality of documents, the classifying comprising;

generating one or more context identifiers for the search request based at least on the identified plurality of documents that include the particular term, wherein each context identifier contains words that occur in context with one or more terms in the search request in one or more of the plurality of documents; and

using the one or more context identifiers, identifying that the one or more terms in the search request and the one or more context identifiers are associated with a media-related request;

modifying the search request comprising

     1) adding one or more terms to the request to include a media-related category, and

     2) discarding one or more words from the search request input by the user, wherein adding the one or more terms to the request to include the media-related category biases search results toward returning media-related results;

generating search results for the modified search request;

formatting the search results in a structured display that is pre-assigned to media related requests by generating markup code for display on the user device, wherein the formatting includes a title and an indication relating to how to access media relating to the title; and

providing, to the user at the user device, the formatted search results for display.

17. The computer-readable medium of claim 16, wherein generating the one or more context identifiers for the search request comprises submitting one or more words in the search request to a learning system to generate a context vector indicative of the one or more words'"'"' relationship to media-related words in a graph.

18. The computer-readable medium of claim 16, wherein generating the one or more context identifiers for the search request based at least on the plurality of documents comprises determining, in a graph of word and word group relationships in a machine learning system, a distance between the one or more terms in the search request and the one or more other terms that appear in a corresponding one of the plurality of documents.

19. The computer-readable medium of claim 16, further comprising generating a list of suggested queries that are semantically related to the search request, and providing the list of suggested queries to the user.