Multimedia search using reshare text on online social networks

• US 10,061,856 B2
• Filed: 01/29/2015
• Issued: 08/28/2018
• Est. Priority Date: 01/29/2015
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Bindal, Anuj, Boucher, Maxime, Thaker, Sahil P., Ghafourian, Mahsa, Jain, Arpit Suresh
• Primary Examiner(s): Nguyen, Cam-Linh
• Application Number:
  US14/609,084
• Publication Number:
  US 20160224672A1
• Time in Patent Office: 1,307 Days
• Field of Search: 707723
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

In one embodiment, a method includes receiving a search query for multimedia objects of an online social network from a user of the online social network; searching an index of multimedia objects to identify multimedia objects indexed with keywords that match n-grams of the search query, the keywords having been extracted from reshares associated with the multimedia objects; calculating an object-score for each identified multimedia object based on social signals; generating search results with references to the identified multimedia objects that have an object-score greater than a threshold object-score; and sending to the user one or more of the search results as part of a search-results page.

Claims

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

receiving, by the one or more computing devices from a client system of a first user of an online social network, a search query to search for multimedia objects of the online social network, the search query comprising one or more n-grams;

searching, by the one or more computing devices, an index of multimedia objects to identify one or more multimedia objects based on the search query, wherein the index of multimedia objects indexes a plurality of multimedia objects and one or more keywords associated with each multimedia object, each keyword being extracted from one or more communications associated with the multimedia object, wherein each communication is of a particular communication-type of a plurality of communication-types, and wherein each communication is sent by one or more second users of the online social network and links to the multimedia object associated with the communication, wherein each identified multimedia object is indexed with one or more keywords matching at least a portion of the one or more n-grams of the search query;

calculating, by the one or more computing devices, for each identified multimedia object, an object-score based on;

(1) one or more social signals associated with the identified multimedia object, (2) whether an author of a communication associated with the identified multimedia object is identified as a key-author determined to be relevant to or knowledgeable about one or more topics associated with the communication, and (3) a communication-type of a communication from which one or more of the matching keywords were sourced, wherein;

if the communication is of a post-communication-type, a first weight is applied to the matching keywords, and

if the communication is of a reshare-communication-type, a second weight is applied to the matching keywords, wherein the second weight is greater than the first weight;

generating, by the one or more computing devices, one or more search results corresponding to one or more of the identified multimedia objects, respectively, each search result comprising a reference to a corresponding identified multimedia object having an object-score greater than a threshold object-score; and

sending, by the one or more computing devices, responsive to the search query, a search-results page to the client system of the first user for display, the search-results page comprising one or more of the generated search results.

2. 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 the first user; and

a plurality of user nodes corresponding to a plurality of second users of the online social network, respectively; and

a plurality of concept nodes corresponding to a plurality of communications of the online social network, each communication linking to a multimedia object of the online social network.

3. The method of claim 1, wherein searching the index of multimedia objects to identify one or more multimedia objects based on the search query comprises:

accessing the index of multimedia objects, the index of multimedia objects comprising a plurality of entries, each entry comprising a link to a multimedia object and one or more keywords extracted from a post, reshare, comment, or message comprising a link to the multimedia object;

identifying one or more posts, reshares, comments, or messages linking to one or more multimedia objects based on matching the one or more n-grams of the search query with the one or more keywords extracted from the identified one or more posts, reshares, comments, or messages; and

identifying the one or more multimedia objects linked by the one or more posts, reshares, comments, or messages.

4. The method of claim 1, wherein the object-score is further based on a quality of the match between the one or more n-grams of the search query and the one or more keywords associated with the identified multimedia object.

5. The method of claim 1, wherein the object-score is further based on a relationship within the online social network between the first user and an author of a communication corresponding to the identified multimedia object.

6. The method of claim 1, wherein the object-score is further based on a current popularity of one or more topics associated with one or more communications corresponding to the identified multimedia object.

7. The method of claim 1, wherein the object-score is further based on a number of times the identified multimedia object has been accessed by users of the online social network.

8. The method of claim 1, wherein the search results of the search-results page are presented in an order based on the object-scores of the identified multimedia objects corresponding to the search results.

9. The method of claim 8, wherein the order is further based on a post-score of a post or reshare, the post-score being based on one or more social signals.

10. The method of claim 9, wherein the post-score is further based on a relationship within the online social network between the first user and an author of the post or reshare.

11. The method of claim 1, wherein the one or more keywords associated with each multimedia object further comprise keywords extracted from one or more comments associated with a reshare associated with the multimedia object.

12. The method of claim 1, wherein the one or more keywords associated with each multimedia object further comprise keywords extracted from one or more comments associated with a post associated with the multimedia object.

13. The method of claim 1, wherein the one or more keywords associated with each multimedia object have been extracted from the communications associated with the multimedia object based on a TF-IDF analysis.

14. The method of claim 1, further comprising:

receiving a first post or reshare comprising a link to a multimedia object and associated text;

accessing a post index, wherein the post index is a forward index;

extracting one or more keywords from the associated text of the first post or reshare; and

indexing the first post or reshare by a post ID in an index entry, the index entry comprising the link and the one or more extracted keywords from the associated text of the first post or reshare.

15. The method of claim 1, further comprising:

receiving a first post or reshare comprising a link to a multimedia object and associated text;

accessing a multimedia index, wherein the multimedia index is an inverted index;

extracting one or more keywords from the associated text of the first post or reshare; and

indexing the link by a link ID in an index entry, the index entry further comprising one or more posts or reshares associated with the link and one or more extracted keywords from associated text of one or more posts or reshares associated with the link.

16. The method of claim 1, further comprising:

receiving a first post or reshare comprising a link to a multimedia object and associated text;

accessing a keyword index, wherein the keyword index is an inverted index;

extracting one or more keywords from the associated text of the first post or reshare; and

indexing the link by an extracted keyword from associated text of one or more posts or reshares associated with the link in an index entry, the index entry further comprising one or more links to one or more multimedia objects associated with the keyword.

17. The method of claim 1, wherein the search query is received by the one or more computing devices following input of the search query into a query field of an interface being displayed on the client system of the first user, wherein the interface is associated with the online social network.

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

receive, by one or more computing devices from a client system of a first user of an online social network, a search query to search for multimedia objects of the online social network, the search query comprising one or more n-grams;

search, by the one or more computing devices, an index of multimedia objects to identify one or more multimedia objects based on the search query, wherein the index of multimedia objects indexes a plurality of multimedia objects and one or more keywords associated with each multimedia object, each keyword being extracted from one or more communications associated with the multimedia object, wherein each communication is of a particular communication-type of a plurality of communication-types, and wherein each communication is sent by one or more second users of the online social network and links to the multimedia object associated with the communication, wherein each identified multimedia object is indexed with one or more keywords matching at least a portion of the one or more n-grams of the search query;

calculate, by the one or more computing devices, for each identified multimedia object, an object-score based on;

(1) one or more social signals associated with the identified multimedia object, (2) whether an author of a communication associated with the identified multimedia object is identified as a key-author determined to be relevant to or knowledgeable about one or more topics associated with the communication, and (3) a communication-type of a communication from which one or more of the matching keywords were sourced, wherein;

if the communication is of a post-communication-type, a first weight is applied to the matching keywords, and

if the communication is of a reshare-communication-type, a second weight is applied to the matching keywords, wherein the second weight is greater than the first weight;

generate, by the one or more computing devices, one or more search results corresponding to one or more of the identified multimedia objects, respectively, each search result comprising a reference to the corresponding identified multimedia object having an object-score greater than a threshold object-score; and

send, by the one or more computing devices, responsive to the search query, a search-results page to the client system of the first user for display, the search-results page comprising one or more of the generated search results.

19. The media of claim 18, wherein the software is further operable when executed to:

access 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 the first user; and

a plurality of user nodes corresponding to a plurality of second users of the online social network, respectively; and

a plurality of concept nodes corresponding to a plurality of communications of the online social network, each communication linking to a multimedia object of the online social network.

20. The media of claim 18, wherein the software when executed to search the index of multimedia objects to identify one or more multimedia objects based on the search query is further operable when executed to:

access the index of multimedia objects, the index of multimedia objects comprising a plurality of entries, each entry comprising a link to a multimedia object and one or more keywords extracted from a post, reshare, comment, or message comprising a link to the multimedia object;

identify one or more posts, reshares, comments, or messages linking to one or more multimedia objects based on matching the one or more n-grams of the search query with the one or more keywords extracted from the identified one or more posts, reshares, comments, or messages; and

identify the one or more multimedia objects linked by the one or more posts, reshares, comments, or messages.

21. The media of claim 18, wherein the object-score is further based on a quality of the match between the one or more n-grams of the search query and the one or more keywords associated with the identified multimedia object.

22. The media of claim 18, wherein the object-score is further based on a relationship within the online social network between the first user and an author of a communication corresponding to the identified multimedia object.

23. The media of claim 18, wherein the object-score is further based on a current popularity of one or more topics associated with one or more communications corresponding to the identified multimedia object.

24. The media of claim 18, wherein the object-score is further based on a number of times the identified multimedia object has been accessed by users of the online social network.

25. The media of claim 18, wherein the search results of the search-results page are presented in an order based on the object-scores of the identified multimedia objects corresponding to the search results.

26. The media of claim 25, wherein the order is further based on a post-score of a post or reshare, the post-score being based on one or more social signals.

27. The media of claim 26, wherein the post-score is further based on a relationship within the online social network between the first user and an author of the post or reshare.

28. The media of claim 18, wherein the one or more keywords associated with each multimedia object further comprise keywords extracted from one or more comments associated with a reshare associated with the multimedia object.

29. The media of claim 18, wherein the one or more keywords associated with each multimedia object further comprise keywords extracted from one or more comments associated with a post associated with the multimedia object.

30. The media of claim 18, wherein the one or more keywords associated with each multimedia object have been extracted from the communications associated with the multimedia object based on a TF-IDF analysis.

31. The media of claim 18, the software is further operable when executed to:

receive a first post or reshare comprising a link to a multimedia object and associated text;

access a post index, wherein the post index is a forward index;

extract one or more keywords from the associated text of the first post or reshare; and

index the first post or reshare by a post ID in an index entry, the index entry comprising the link and the one or more extracted keywords from the associated text of the first post or reshare.

32. The media of claim 18, the software is further operable when executed to:

receive a first post or reshare comprising a link to a multimedia object and associated text;

access a multimedia index, wherein the multimedia index is an inverted index;

extract one or more keywords from the associated text of the first post or reshare; and

index the link by a link ID in an index entry, the index entry further comprising one or more posts or reshares associated with the link and one or more extracted keywords from associated text of one or more posts or reshares associated with the link.

33. The media of claim 18, further comprising:

receive a first post or reshare comprising a link to a multimedia object and associated text;

access a keyword index, wherein the keyword index is an inverted index;

extract one or more keywords from the associated text of the first post or reshare; and

index the link by an extracted keyword from associated text of one or more posts or reshares associated with the link in an index entry, the index entry further comprising one or more links to one or more multimedia objects associated with the keyword.

34. 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, by one or more of the processors from a client system of a first user of an online social network, a search query to search for multimedia objects of the online social network, the search query comprising one or more n-grams;

search, by one or more of the processors, an index of multimedia objects to identify one or more multimedia objects based on the search query, wherein the index of multimedia objects indexes a plurality of multimedia objects and one or more keywords associated with each multimedia object, each keyword being extracted from one or more communications associated with the multimedia object, wherein each communication is of a particular communication-type of a plurality of communication-types, and wherein each communication is sent by one or more second users of the online social network and links to the multimedia object associated with the communication, wherein each identified multimedia object is indexed with one or more keywords matching at least a portion of the one or more n-grams of the search query;

calculate, by one or more of the processors, for each identified multimedia object, an object-score based on;

(1) one or more social signals associated with the identified multimedia object, (2) whether an author of a communication associated with the identified multimedia object is identified as a key-author determined to be relevant to or knowledgeable about one or more topics associated with the communication, and (3) a communication-type of a communication from which one or more of the matching keywords were sourced, wherein;

if the communication is of a post-communication-type, a first weight is applied to the matching keywords, and

if the communication is of a reshare-communication-type, a second weight is applied to the matching keywords, wherein the second weight is greater than the first weight;

generate, by one or more of the processors, one or more search results corresponding to one or more of the identified multimedia objects, respectively, each search result comprising a reference to the corresponding identified multimedia object having an object-score greater than a threshold object-score; and

send, by one or more of the processors, responsive to the search query, a search-results page to the client system of the first user for display, the search-results page comprising one or more of the generated search results.