Providing objective and people results for search

• US 8,538,960 B2
• Filed: 08/05/2011
• Issued: 09/17/2013
• Est. Priority Date: 08/05/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wong, Sandy, Mu, Wei, Yin, Jun, Nair, Rahul, King, Simon, Badrinarayanan, Srinivasan, Legros, Xavier, Ching, Michael, Haas, Kevin, Nabar, Shubha
• Primary Examiner(s): LEROUX, ETIENNE PIERRE
• Application Number:
  US13/204,631
• Publication Number:
  US 20130036114A1
• Time in Patent Office: 774 Days
• Field of Search: 707/2, 707/3, 707/5, 707/732, 715/751, 379/142.05
• US Class Current: 707/732
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

Search results may include both objective results and person results. In one example, a search query is evaluated to determine whether it is the type of query that a user might want to ask to a friend. If the query is of such a type, then the search engine may examine a social graph to determine which friends of the user who entered the query may have information that is relevant to answering the query. If such friends exist, then the friends may be displayed along with objective search results, along with an explanation of each friend'"'"'s relevance to the query. Clicking on a person in the results may cause a conversation to be initiated with that person, thereby allowing the user who entered the query to ask his or her friend about the subject of the query.

Claims

1. One or more computer-readable storage memories that store executable instructions to provide search results, the executable instructions, when executed by a computer, causing the computer to perform acts comprising:

receiving a query from a user;

determining that the query is answerable with subjective or socially-derived information;

comparing said query to a corpus of information to obtain objective results;

comparing said query to a social graph to identify one or more people whose relationship to said user meets a closeness condition and who have an aspect of relevance to said query;

creating person results that comprise said one or more people and, for each of said one or more people, an explanation of each person'"'"'s relevance to said query;

providing, to said user, a set of results that comprise said objective results and said person results; and

training a classifier to identify queries that call for subjective information using training data that comprises;

a plurality of positive examples in which people were provided as search results and in which users who requested the results clicked on the people in the results; and

a plurality of negative examples in which people were provided as search results and in which users who requested the results did not click on the people in the results;

said determining that said query calls for subjective information being performed using said classifier, with said classifier having been trained on said training data.

2. The one or more computer-readable storage memories of claim 1, said aspect of relevance being based on a comparison of words in said query with attributes of people in said social graph.

3. The one or more computer-readable storage memories of claim 1, said aspect of relevance being based on a comparison of words in said query with things or concepts for which people in said social graph have declared affinity.

4. The one or more computer-readable storage memories of claim 1, said query indicating a first geographic location, said aspect of relevance being based on a comparison of said first geographic locations with geographic locations of people in said social graph, said query being found to match a person who is associated with a second geographic location different from said first geographic location based on distance between said first geographic location and said second geographic location.

5. The one or more computer-readable storage memories of claim 1, said aspect of relevance being based on a comparison of words in said query with a characteristic that is derived from likes or text posts of a person in said social graph, said query being found to match said person based on words in said query matching said characteristic.

6. The one or more computer-readable storage memories of claim 1, said aspect of relevance being based on a determination that words in said query and an annotation of a person in said social graph are both associated with a concept in a concept graph.

7. The one or more computer-readable storage memories of claim 1, said determining that said query is answerable with subjective or socially-derived information being performed after a comparison of said query to said social graph, said determining being based on whether there are a sufficient number of people in said social graph who meet said closeness condition and have said aspect of relevance.

8. The one or more computer-readable storage memories of claim 1, said determining that said query is answerable with subjective or socially-derived information being based on specific words being in said query.

9. The one or more computer-readable storage memories of claim 1, said person results being provided to said user in a form such that clicking a person in said person results initiates a conversation between said user and said person.

10. A method of providing search results, the method comprising:

using a processor to perform acts comprising;

receiving a query from a user;

determining that the query is answerable with subjective information;

obtaining objective results from a corpus of information;

using a social graph to identify one or more people who are friends of said user in said social graph, and who have an aspect of relevance to said query;

creating person results that comprise said one or more people and, for each of said one or more people, an explanation of each person'"'"'s relevance to said query;

providing, to said user, a set of results that comprise said objective results and said person results, each of said person results being actionable to initiate contact between said user and a person identified in a given result; and

training a classifier to identify queries that call for subjective information using training data that comprises;

a plurality of positive examples of queries for which people were provided as search results and in which users who requested the results clicked on the people in the results; and

a plurality of negative examples of queries for which people were provided as search results and in which users who requested the results did not click on the people in the results;

said determining that said query calls for subjective information being performed using said classifier, with said classifier having been trained on said training data.

11. The method of claim 10, said aspect of relevance being based on a comparison of words in said query with wall posts or tweets of people in said social graph.

12. The method of claim 10, said aspect of relevance being based on a comparison of words in said query with Uniform Resource Locators that people in said social graph have entered.

13. The method of claim 10, said query indicating a first geographic location, said aspect of relevance being based on a comparison of said first geographic locations with geographic locations of people in said social graph, said query being found to match a person who is associated with a second geographic location different from said first geographic location based on distance between said first geographic location and said second geographic location.

14. The method of claim 10, said determining that said query is answerable with subjective information being performed after a comparison of said query to said social graph, said determining being based on whether there are a sufficient number of people in said social graph who are friends with said user and who have said aspect of relevance.

15. The method of claim 10, said determining that said query is answerable with subjective information being based said query'"'"'s containing specific words that have been determined to be associated with subjective requests.

16. A system for providing search results, the system comprising:

a memory;

a processor; and

a component that is stored in said memory, that executes on said processor, that receives a query from a user, that determines that said query is answerable, in part, with subjective information, that obtains objective results corresponding to the query from a corpus of information, that identifies one or more people in a social graph whose relationship to said user meets a closeness condition and who have an aspect of relevance to said query, that creates person results that comprise a portion of said one or more people, that creates, for each of said one or more people, an explanation of each person'"'"'s relevance to said query, and that provides, to said user, a set of results that comprise said objective results and said person results;

a classifier that is trained to identify queries that are answerable with subjective information, said classifier being trained on training data that comprises;

a plurality of positive examples in which people were provided as search results and in which users who requested the results clicked on the people in the results; and

a plurality of negative examples in which people were provided as search results and in which users who requested the results did not click on the people in the results;

said component using said classifier to determine that said query is answerable, in part, with subjective information.

17. The system of claim 16, said aspect of relevance being based on a comparison of words in said query with attributes of people in said social graph.

18. The system of claim 16, said aspect of relevance being based on a comparison of words in said query with ads that people in said social graph have clicked-through.

19. The system of claim 16, said aspect of relevance being based on a comparison of words in said query with a characteristic that is derived from declared affinities or text posts of a person in said social graph, said query being found to match said person based on words in said query matching said characteristic.

20. The system of claim 16, said aspect of relevance being based on a determination that words in said query and an annotation of a person in said social graph are both associated with a concept in a concept graph.

21. The system of claim 16, wherein said component determines that said query is answerable, in part, with subjective information after said component compares said query to said social graph, and wherein said component determines that said query is answerable, in part, with subjective information based on whether there are a sufficient number of people in said social graph who meet said closeness condition and have said aspect of relevance.

22. The system of claim 16, wherein said component determines that said query is answerable, in part, with subjective information based on whether specific words appear in said query.

23. The system of claim 16, said person results being provided to said user in a form such that clicking a person in said person results initiates a conversation between said user and said person.