Vital author snippets on online social networks

Patent metadata

Field	Value
Publication number	US-10614141-B2
Application number	US-201715459678-A
Country	US
Kind code	B2
Filing date	Mar 15, 2017
Priority date	Mar 15, 2017
Publication date	Apr 7, 2020
Grant date	Apr 7, 2020

Abstract

In one embodiment, a method includes receiving a text query from a client system of a user and parsing the text query to identify a primary entity referenced in the text query. The method also includes identifying one or more related entities for the primary entity based on one or more related-entity indexes associated with the primary entity and identifying one or more content objects matching the text query, each identified content object being associated with one or more of the related entities. The method also includes sending to the client system instructions for presenting one or more search results corresponding to one or more of the identified content objects, respectively, each search result including a reference to the associated related entity and a snippet for the related entity describing the relationship between the primary entity and the related entity.

First claim

What is claimed is:

1. A method comprising, by one or more computing devices:
identifying, from a plurality of entities, a primary entity and one or more related entities for the primary entity based on one or more related-entity indexes associated with the primary entity, wherein each related-entity index for the primary entity comprises identification information of one or more related entities to the primary entity;
identifying, for each identified related entity, one or more content objects associated with the related entity and the primary entity;
extracting, for each identified related entity, structured data from the identified content objects associated with the related entity, wherein the structured data indicates a relationship between the primary entity and the related entity;
generating, for each identified related entity, a snippet for the related entity based on an application of one or more snippet templates to the structured data extracted from the identified content objects associated with the related entity, wherein each snippet describes the relationship between the primary entity and the related entity; and
storing, for each identified related entity, the generated snippet in association with the identification information of the related entity.

2. The method of claim 1 , further comprising:
receiving a text query comprising one or more n-grams inputted by a first user.

3. The method of claim 1 , wherein one or more of the related-entity indexes associated with the primary entity are generated based at least in part on one or more online encyclopedic indexes.

4. The method of claim 1 , wherein one or more of the related-entity indexes associated with the primary entity are generated based at least in part on social-graph information associated with the primary entity.

5. The method of claim 1 , wherein one or more of the related-entity indexes associated with the primary entity are generated based at least in part on a co-occurrence of references to the primary entity with references to one or more of the related entities within one or more content objects associated with the primary entity and one or more of the related entities.

6. The method of claim 1 , further comprising:
generating one or more of the related-entity indexes associated with the primary entity by:
identifying one or more similar entities to the primary entity; and
populating the one or more generated related-entity indexes associated with the primary entity with one or more related entities extracted from one or more related-entity indexes associated with the identified similar entities.

7. The method of claim 1 , wherein the snippet generated for each of the identified related entities comprises a number of characters less than or equal to a specified character limit.

8. The method of claim 1 , wherein, for one or more of the identified related entities, at least one of the identified content objects is published on a profile interface corresponding to the identified related entity.

9. The method of claim 1 , wherein the extracted structured data comprises one or more text strings, and wherein generating the snippet based on the application of one or more snippet templates to the structured data extracted from the identified content objects associated with the related entity comprises one or more of:
removing one or more phrases from one or more of the text strings;
adding one or more phrases to one or more of the text strings;
changing an order of one or more phrases within one or more of the text strings; or
combining one or more of the text strings.

10. The method of claim 1 , wherein generating the snippet based on the application of one or more snippet templates to the structured data extracted from the identified content objects associated with the related entity comprises analyzing the extracted structured data using a natural-language processing algorithm.

11. The method of claim 1 , wherein generating the snippet is based on an application of one or more pre-generated snippet templates to the structured data extracted from the identified content objects associated with the related entity.

12. The method of claim 1 , further comprising, for each identified related entity calculating a confidence score for the generated snippet based at least in part on one or more of:
a measure of reliability of the content objects associated with the related entity, or
a measure of strength of a relationship between the primary entity and the related entity described by the snippet; and
storing, for each identified related entity, the generated snippet having the greatest confidence score in association with the identification information of the related entity.

13. The method of claim 1 , further comprising, for one or more of the identified related entities:
calculating a score for each of the identified content objects based at least in part on a number of social signals associated with:
the content object, and
one or more related entities associated with the content object; and

generating, responsive to receiving a text query comprising one or more n-grams inputted by a first user, one or more search results comprising references to one or more of the identified content objects, respectively, each having a calculated score greater than a threshold score.

14. The method of claim 13 , wherein the search results are presented to the first user in a ranked order based at least in part on the respective scores of the content objects referenced by the search results.

15. The method of claim 13 , wherein the social signals associated with the content object comprise one or more of:
a number of likes associated with the content object;
a number of comments on the content object;
a number of shares of the content object; or
a change in a growth rate of an amount of social activities associated with the content object.

16. The method of claim 13 , wherein the social signals associated with a related entity 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 associated with the related entity.

17. The method of claim 13 , wherein calculating the score for each of the identified content objects comprises:
determining a first meaning of one of the n-grams of the text query;
determining a second meaning of an n-gram within the context of the content object; and
calculating the score based at least in part on a level of matching between the first meaning and the second meaning.

18. 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:
identify, from a plurality of entities, a primary entity and one or more related entities for the primary entity based on one or more related-entity indexes associated with the primary entity, wherein each related-entity index for the primary entity comprises identification information of one or more related entities to the primary entity;
identify, for each identified related entity, one or more content objects associated with the related entity and the primary entity;
extract, for each identified related entity, structured data from the identified content objects associated with the related entity, wherein the structured data indicates a relationship between the primary entity and the related entity;
generate, for each identified rel

Assignees

• Facebook Inc

Inventors

• Pawar Manoj Mahipat
• Mashiach Li-Tal

Classifications

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• G06F16/24578

  using ranking · CPC title

• G06F16/248

  Presentation of query results · CPC title

• G06F16/2228

  Indexing structures · CPC title