Social search engine

• US 9,098,808 B1
• Filed: 04/21/2011
• Issued: 08/04/2015
• Est. Priority Date: 04/22/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Ventilla, Max, Spiro, Robert J., Horowitz, Damon, Kamvar, Sepandar D., MacCartney, William C., Davies, Winton Henry Ellis, Kim, Michael
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Sitiriche, Luis
• Application Number:
  US13/092,024
• Time in Patent Office: 1,566 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Methods, systems, and apparatus, including computer programs, for a social search engine. In one aspect, a method includes receiving an answer to a question from an answerer; identifying an entity in the answer, in which the entity refers to a product, a service, a company, or a merchant; obtaining permission from the answerer to provide an interactive link for the entity in the answer to a resource from which the entity can be obtained; receiving a user submitted query and determining that the query is related to the entity; and responsive to determining, providing the answer, including the link to the resource, as part of search results that are responsive to the query.

Claims

1. A method performed by data processing apparatus, the method comprising:

receiving a query from an asker;

before routing the query to at least one candidate answerer;

calculating for each of a plurality of users a respective first probability that the user will successfully answer a query by an asker about one or more topics, wherein the first probability is based at least partly on topics in which the user has expertise, and, for at least one user, the first probability being based at least partly on an additional topic that is not explicitly associated with the at least one user,

calculating for each of the plurality of users a respective query-independent second probability that the user will provide a satisfactory answer to the asker, wherein the second probability is based at least partly on a degree of social indirection between the user and the asker,

determining a score for each of the users based at least in part on their respective first and second probabilities, and

defining a list of candidate answerers based on ranking users of the plurality of users based on respective scores, the list of candidate answerers comprising one or more candidate answerers; and

routing the query to the one or more candidate answerers of the list of candidate answerers.

2. The method of claim 1 wherein the plurality of users belong to a social affinity group.

3. The method of claim 1 wherein calculating for each of the plurality of users the respective query-independent second probability that the user will provide a satisfactory answer to the asker is based partly on one or more of the following factors between the asker and the user:

social connection similarity, demographic similarity, profile similarity, vocabulary similarity, chattiness similarity, verbosity similarity, politeness similarity, and responsiveness similarity.

4. The method of claim 1, further comprising ranking users of the plurality of users based on the determined scores, the one or more candidate answerers of the list of candidate answerers including highest ranked users.

5. The method of claim 1 wherein the query is location-sensitive, and wherein the plurality of users is associated with the location.

6. The method of claim 1, further comprising:

associating one or more topics with the user by analyzing text associated with the user and with others in a social affinity group of the user; and

determining a probability distribution for the topics based on expertise in the topics of the user and the others.

7. The method of claim 6 wherein the text is one or more:

topics provided by the user, topics provided by the others in the social affinity group, the user'"'"'s online profile, text in web pages associated with the user, status message updates sent by the user, and messages sent to/from the others.

8. The method of claim 6 wherein each topic in the one or more topics has a score depending on a confidence for a source of the topic.

9. The method of claim 6, further comprising disassociating a topic in the one or more topics with the user if one or more of the following conditions holds true:

the user mutes the topic, the user declines to answer a question about the topic, and the user receives negative feedback on an answer about the topic from another user.

10. A system comprising:

a storage medium having instructions stored thereon; and

data processing apparatus operable to execute the instructions to perform operations comprising;

receiving a query from an asker;

before routing the query to at least one candidate answerer;

calculating for each of a plurality of users a respective first probability that the user will successfully answer a query by an asker about one or more topics, wherein the first probability is based at least partly on topics in which the user has expertise, and, for at least one user, the first probability being based at least partly on an additional topic that is not explicitly associated with the at least one user,

calculating for each of the plurality of users a respective query-independent second probability that the user will provide a satisfactory answer to the asker, wherein the second probability is based at least partly on a degree of social indirection between the user and the asker,

determining a score for each of the users based at least in part on their respective first and second probabilities, and

defining a list of candidate answerers based on ranking users of the plurality of users based on respective scores, the list of candidate answerers comprising one or more candidate answerers; and

routing the query to the one or more candidate answerers of the list of candidate answerers.

11. The system of claim 10 wherein the plurality of users belong to a social affinity group.

12. The system of claim 10 wherein calculating for each of the plurality of users the respective query-independent second probability that the user will provide a satisfactory answer to the asker is based partly on one or more of the following factors for the asker and user pair:

social connection similarity, demographic similarity, profile similarity, vocabulary similarity, chattiness similarity, verbosity similarity, politeness similarity, and responsiveness similarity.

13. The system of claim 10 wherein the operations further comprise ranking users of the plurality of users based on the determined scores, the one or more candidate answerers of the list of candidate answerers including highest ranked users.

14. The system of claim 10 wherein the query is location-sensitive, and wherein the plurality of users is associated with the location.

15. The system of claim 10 wherein the operations further comprise:

associating one or more topics with the user by analyzing text associated with the user and with others in a social affinity group of the user; and

determining a probability distribution for the topics based on expertise in the topics of the user and the others.

16. The system of claim 15 wherein the text is one or more:

topics provided by the user, topics provided by the others in the social affinity group, the user'"'"'s online profile, text in web pages associated with the user, status message updates sent by the user, and messages sent to/from the others.

17. The system of claim 15 wherein each topic in the one or more topics has a score depending on a confidence for a source of the topic.

18. The system of claim 15 wherein the operations further comprise disassociating a topic in the one or more topics with the user if one or more of the following conditions holds true:

the user mutes the topic, the user declines to answer a question about the topic, and the user receives negative feedback on an answer about the topic from another user.

19. A non-transitory storage medium having instructions stored thereon that, when executed by data processing apparatus, cause the data processing apparatus to perform operations comprising:

receiving a query from an asker;

before routing the query to at least one candidate answerer;

calculating for each of a plurality of users a respective first probability that the user will successfully answer a query by an asker about one or more topics, wherein the first probability is based at least partly on topics in which the user has expertise, and, for at least one user, the first probability being based at least partly on an additional topic that is not explicitly associated with the at least one user,

calculating for each of the plurality of users a respective query-independent second probability that the user will provide a satisfactory answer to the asker, wherein the second probability is based at least partly on a degree of social indirection between the user and the asker,

determining a score for each of the users based at least in part on their respective first and second probabilities, and

defining a list of candidate answerers based on ranking users of the plurality of users based on respective scores, the list of candidate answerers comprising one or more candidate answerers; and

routing the query to the one or more candidate answerers of the list of candidate answerers.

20. The storage medium of claim 19 wherein the plurality of users belong to a social affinity group.

21. The storage medium of claim 19 wherein calculating for each of the plurality of users the respective query-independent second probability that the user will provide a satisfactory answer to the asker is based partly on one or more of the following factors for the asker and user pair:

social connection similarity, demographic similarity, profile similarity, vocabulary similarity, chattiness similarity, verbosity similarity, politeness similarity, and responsiveness similarity.

22. The storage medium of claim 19 wherein the operations further comprise ranking users of the plurality of users based on the determined scores, the one or more candidate answerers of the list of candidate answerers including highest ranked users.

23. The storage medium of claim 19 wherein the query is location-sensitive, and wherein the plurality of users is associated with the location.

24. The storage medium of claim 19 wherein the operations further comprise:

associating one or more topics with the user by analyzing text associated with the user and with others in a social affinity group of the user; and

determining a probability distribution for the topics based on expertise in the topics of the user and the others.

25. The storage medium of claim 24 wherein the text is one or more:

topics provided by the user, topics provided by the others in the social affinity group, the user'"'"'s online profile, text in web pages associated with the user, status message updates sent by the user, and messages sent to/from the others.

26. The storage medium of claim 24 wherein each topic in the one or more topics has a score depending on a confidence for a source of the topic.

27. The storage medium of claim 24 wherein the operations further comprise disassociating a topic in the one or more topics with the user if one or more of the following conditions holds true:

the user mutes the topic, the user declines to answer a question about the topic, and the user receives negative feedback on an answer about the topic from another user.