Generating Answers to Questions Using Information Posted By Users on Online Social Networks

• US 20170206271A1
• Filed: 01/20/2016
• Published: 07/20/2017
• Est. Priority Date: 01/20/2016
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Jain, Arpit Suresh
• Granted Patent: US 10,740,375 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

In one embodiment, a method includes A method comprising, by one or more computing devices of an online social network, receiving, from a client system of a first user of the online social network, a post that is parsed to identify a query, identifying one or more topics and one or more query-domains matching the query, searching an answers-index to identify one or more query-answer sets responsive to the query and matching at least one of the identified topics and at least one of the identified domains, searching, for each entity identified in each query-answer set, a posts-index to identify one or more comments corresponding to the identified entity, calculating a score for each comment based on social signals associated with the comment, and sending, to the client system of the first user, a suggested-answers page including references to each comment having a score greater than a threshold score.

Claims

1. A method comprising, by one or more computing devices of an online social network:

receiving, from a client system of a first user of the online social network, a post inputted by the first user, wherein the post is parsed to identify a query associated with the post;

identifying one or more topics and one or more query-domains matching the query;

searching an answers-index to identify one or more query-answer sets responsive to the query, each query-answer set matching at least one of the identified topics and at least one of the identified domains, wherein each query-answer set identifies one or more entities;

searching, for each entity identified in each query-answer set, a posts-index to identify one or more comments corresponding to the identified entity;

calculating a score for each comment based at least on a number of social signals associated with the comment; and

sending, to the client system of the first user, a suggested-answers page comprising references to each comment having a score greater than a threshold score.

2. The method of claim 1, further comprising determining, using natural-language processing, that the query is a question in a natural language syntax.

3. The method of claim 1, wherein identifying the one or more topics of the query comprises using a topic tagger to identify the topics based on the query.

4. The method of claim 1, wherein the query-domains are identified by searching a knowledge base that associates topics with query-domains.

5. The method of claim 1, wherein receiving, from a client system of a first user of the online social network, a query comprises receiving a post that includes the query.

6. The method of claim 1, further comprising:

accessing a social graph comprising a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between them, the nodes comprising;

a first node corresponding to a first user associated with an online social network; and

a plurality of second nodes that each correspond to a concept or a second user associated with the online social network.

7. The method of claim 1, the suggested-answers page further comprising a reference to a post to which the at least one of the identified comments is a response.

8. The method of claim 7, wherein the suggested-answers page comprises an expandable thread user interface for at least one of the identified comments, and the expandable thread user interface is configured to display the post and the at least one of the identified comments in response to user selection of an expand command.

9. The method of claim 1, wherein each query-answer set matches all of the identified topics and identified domains.

10. The method of claim 1, further comprising ranking the comments based on the score associated with each comment.

11. The method of claim 10, wherein the score for each comment is further based on how many of the identified topics and identified domains are matched by the query-answer set that identifies the entity corresponding to the comment.

12. The method of claim 1, wherein the social signals comprise author information or location information associated with each comment.

13. The method of claim 1, wherein the social signals comprise affinity of the first user for each comment or affinity of the first user for an author of each comment.

14. The method of claim 1, further comprising:

identifying a post by a second user that comprises a query;

identifying one or more query topics of the query;

identifying one or more query-domains of each of the query topics;

identifying one or more comments associated with the post by the second user, wherein the comments each comprise an answer topic; and

generating, in the answers-index, one or more answers-index entries that associate each of the domains and corresponding question topics with the answer topic.

15. The method of claim 14, further comprising identifying an answer entity that corresponds to the answer topic by searching a social graph for the answer entity, wherein a name of the answer entity corresponds to the answer topic, and the one or more answers-index entries associate each of the domains and corresponding question topics with the answer entity.

16. The method of claim 14, further comprising:

identifying one or more answer-domains of each answer topic; and

determining that the query-domains match the answer-domains,

wherein the generating, in the answers-index, the one or more answers-index entries is conditional upon the query-domains of each query topic matching the answer-domains of each answer topic.

17. The method of claim 14, further comprising generating, in the posts-index, a posts-index entry that associates the answer topic with each of the one or more comments that comprise the answer topic.

18. The method of claim 17, wherein the answer topic corresponds to an entity, and the posts-index entry associates the corresponding entity with each of the one or more comments that comprise the answer topic.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

receive, from a client system of a first user of the online social network, a post inputted by the first user, wherein the post is parsed to identify a query associated with the post;

identify one or more topics and one or more query-domains matching the query;

search an answers-index to identify one or more query-answer sets responsive to the query, each query-answer set matching at least one of the identified topics and at least one of the identified domains, wherein each query-answer set identifies one or more entities;

search, for each entity identified in each query-answer set, a posts-index to identify one or more comments corresponding to the identified entity;

calculate a score for each comment based at least on a number of social signals associated with the comment; and

send, to the client system of the first user, a suggested-answers page comprising references to each comment having a score greater than a threshold score.

20. A system comprising:

one or more processors; and

a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to;

receive, from a client system of a first user of the online social network, a post inputted by the first user, wherein the post is parsed to identify a query associated with the post;

identify one or more topics and one or more query-domains matching the query;

search an answers-index to identify one or more query-answer sets responsive to the query, each query-answer set matching at least one of the identified topics and at least one of the identified domains, wherein each query-answer set identifies one or more entities;

search, for each entity identified in each query-answer set, a posts-index to identify one or more comments corresponding to the identified entity;

calculate a score for each comment based at least on a number of social signals associated with the comment; and

send, to the client system of the first user, a suggested-answers page comprising references to each comment having a score greater than a threshold score.