Searching for posts by related entities on online social networks

• US 10,235,469 B2
• Filed: 11/30/2016
• Issued: 03/19/2019
• Est. Priority Date: 11/30/2016
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Ahuja, Gaurav
• Primary Examiner(s): Casanova, Jorge A
• Application Number:
  US15/365,113
• Publication Number:
  US 20180150559A1
• Time in Patent Office: 839 Days
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

In one embodiment, a method includes receiving a text query comprising one or more n-grams inputted by a first user, identifying one or more primary entities matching one or more n-grams of the received text query, and identifying, for each of the identified primary entities, one or more related entities based on one or more related-entity indexes associated with the primary entity. The method also includes accessing, for each identified related entity, one or more posts authored by the identified related entity, each accessed post matching all of the n-grams of the received text query, calculating a score for each of the accessed posts, and sending to the first user one or more search results corresponding to one or more of the accessed posts having scores higher than a threshold score, respectively.

Claims

1. A method comprising, by one or more computing devices:

receiving, from a client system of a first user of an online social network, a text query comprising one or more n-grams inputted by the first user;

identifying one or more primary entities matching one or more n-grams of the received text query;

identifying, for each of the identified primary entities, one or more related entities based on one or more related-entity indexes associated with the primary entity, wherein each related-entity index maps from identification information of the primary entity to identification information of related entities to the primary entity;

accessing, for each identified related entity, one or more posts authored by the identified related entity, each accessed post matching all of the n-grams of the received text query;

calculating a score for each of the accessed posts based at least in part on a number of social signals associated with the post and the related entity authoring the post; and

sending, to the client system of the first user for display in response to the first user inputting the one or more n-grams of the text query, one or more search results corresponding to one or more of the accessed posts having scores higher than a threshold score, respectively.

2. The method of claim 1, further comprising:

generating one or more of the related-entity indexes based at least in part on one or more online encyclopedic indexes.

3. The method of claim 1, further comprising:

generating one or more of the related-entity indexes based at least in part on social-graph information associated with the one or more primary entities.

4. The method of claim 1, further comprising:

generating one or more of the related-entity indexes based at least in part on a co-occurrence of references to the one or more primary entities with references to the one or more related entities within content associated with the online social network.

5. The method of claim 1, further comprising generating a related-entity index for at least one of the identified primary entities by:

identifying one or more similar entities to the primary entity; and

populating the related-entity index for the primary entity with one or more related entities extracted from one or more related-entity indexes associated with the identified similar entities.

6. The method of claim 1, wherein each related-entity index comprises relevance scores for the related entities with respect to the primary entity associated with the related-entity index.

7. The method of claim 6, further comprising, for each related-entity index associated with a primary entity, calculating the relevance score for each related entity based at least in part on one or more of:

a similarity coefficient between the related entity and the primary entity;

an affinity coefficient between the related entity and the primary entity;

a frequency of co-occurrence of references to the related entity and the primary entity within content associated with the online social network;

or

a related-entity index associated with the related entity listing the primary entity.

8. The method of claim 6, wherein calculating the score for at least one of the accessed posts is further based on a relevance score for the related entity authoring the post with respect to the primary entity.

9. The method of claim 1, wherein one or more of the accessed posts are published on one or more profile interfaces corresponding to one or more of the identified related entities, respectively.

10. The method of claim 1, wherein the social signals associated with the post comprise one or more of:

a number of likes associated with the post;

a number of comments on the post;

a number of shares of the post;

or

a change in a growth rate of an amount of social activities associated with the post.

11. The method of claim 1, wherein the social signals associated with the related entity authoring the post comprise one or more of:

a number of followers for the related entity;

a number of visits to a profile interface corresponding to the related entity;

or

a number of likes on the related entity.

12. The method of claim 1, wherein each search result comprises a snippet associated with the related entity authoring the post corresponding to the search result, wherein the snippet comprises information describing a relationship between the related entity and the primary entity associated with the related entity.

13. The method of claim 1, wherein the search results are organized based at least in part on the scores of the posts associated with the search results.

14. The method of claim 1, further comprising:

accessing, for each identified primary entity, one or more posts authored by the identified primary entity, each accessed post matching all of the n-grams of the received text query;

calculating a score for each of the accessed posts based at least in part on a number of social signals associated with the post and the primary entity authoring the post; and

sending, to the client system of the first user for display in response to the first user inputting the one or more n-grams of the text query, one or more search results corresponding to one or more of the accessed posts having scores higher than a threshold score, respectively.

15. The method of claim 1, further comprising:

storing text associated with each of one or more trending topics included in a list of trending topics, the text comprising one or more of;

a keyword;

a headline describing the topic;

or

a link to a content object about the topic; and

identifying one or more trending topics matching one or more n-grams of the received text query.

16. The method of claim 15, further comprising, for each of the identified trending topics, identifying one or more primary entities matching the stored text associated with the identified trending topic.

17. The method of claim 15, wherein calculating the score for each of the accessed posts is further based on a time decay factor associated with the post, wherein the time decay factor for each post is based on a duration of time between a time associated with the post and a time of an event associated with one of the identified trending topics.

18. The method of claim 15, wherein calculating the score for each of the accessed posts is further based on a geographic distance between a location associated with the related entity authoring the post and a location of an event associated with one of the identified trending topics.

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 an online social network, a text query comprising one or more n-grams inputted by the first user;

identify one or more primary entities matching one or more n-grams of the received text query;

identify, for each of the identified primary entities, one or more related entities based on one or more related-entity indexes associated with the primary entity, wherein each related-entity index maps from identification information of the primary entity to identification information of related entities to the primary entity;

access, for each identified related entity, one or more posts authored by the identified related entity, each accessed post matching all of the n-grams of the received text query;

calculate a score for each of the accessed posts based at least in part on a number of social signals associated with the post and the related entity authoring the post; and

send, to the client system of the first user for display in response to the first user inputting the one or more n-grams of the text query, one or more search results corresponding to one or more of the accessed posts having scores higher than a threshold score, respectively.

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 an online social network, a text query comprising one or more n-grams inputted by the first user;

identify one or more primary entities matching one or more n-grams of the received text query;

identify, for each of the identified primary entities, one or more related entities based on one or more related-entity indexes associated with the primary entity, wherein each related-entity index maps from identification information of the primary entity to identification information of related entities to the primary entity;

access, for each identified related entity, one or more posts authored by the identified related entity, each accessed post matching all of the n-grams of the received text query;

calculate a score for each of the accessed posts based at least in part on a number of social signals associated with the post and the related entity authoring the post; and

send, to the client system of the first user for display in response to the first user inputting the one or more n-grams of the text query, one or more search results corresponding to one or more of the accessed posts having scores higher than a threshold score, respectively.