Systems, methods, and apparatuses for implementing an interface to view and explore socially relevant concepts of an entity graph

• US 10,318,538 B2
• Filed: 08/27/2015
• Issued: 06/11/2019
• Est. Priority Date: 03/14/2013
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Dunn, Jeffrey, Marlow, Cameron, Nandagopal, Venkataramanan, Bohannon, Philip
• Primary Examiner(s): Truong, Dennis
• Application Number:
  US14/837,732
• Publication Number:
  US 20150363475A1
• Time in Patent Office: 1,384 Days
• Field of Search: 707706, 707722, 707736, 707749, 707798, 707748, 707769
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

There are provided means for implementing an interface to view and explore socially relevant concepts of an entity graph including, for example, means of a social network system to perform operations including retrieving contextually relevant data for a plurality of concepts within an entity graph of the social network system; retrieving socially relevant data for a user'"'"'s node within a social graph of the social network system; identifying intersects between the plurality of concepts within the entity graph and the social relevant data for the user'"'"'s node within the social graph; selecting one of the plurality of concepts within the entity graph based on the intersects identified; and displaying the one of the plurality of concepts within the entity graph at a user interface associated with the user'"'"'s node.

Claims

1. A method within a social network system, the method comprising:

receiving, from a user of the social network system, a search request including one or more search parameters;

retrieving contextually relevant data for a plurality of concepts within an entity graph of the social network system based upon the one or more search parameters, wherein the plurality of concepts comprises two or more of places, people, events, or monetization recommendations;

retrieving socially relevant data for the user within a social graph of the social network system by accessing a node for the user in the social graph that includes socially relevant data describing relationships to friend nodes of the user within the social graph and by accessing one or more concept nodes within the entity graph expressly related to the user node within the social graph or to the friend nodes of the user or both;

identifying intersects between the plurality of concepts within the entity graph and the socially relevant data for the user node within the social graph;

selecting sub-concept recommendations for the plurality of concepts within the entity graph based on the intersects identified by;

creating a common dataset from;

(a) entity objects representing the sub-concept recommendations for the plurality of concepts which are not expressly related to the user node, (b) friend objects representing the friend nodes of the user and expressly related to the user node, and (c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node;

scoring entries in the common dataset by applying a TF-IDF (Term Frequency-Inverse Document Frequency) utility to the common dataset; and

selecting the sub-concept recommendations for the plurality of concepts based on the scores of the entries, wherein the sub-concept recommendations comprise of instances of the plurality of concepts; and

dynamically generating, for display on a user interface associated with the user in response to the search request, a concept page view with the selected sub-concept recommendations for the plurality of concepts within the entity graph.

2. The method of claim 1, wherein retrieving contextually relevant data comprises:

(i) retrieving a list of entity objects representing the plurality of concepts within the entity graph of the social network system;

(ii) filtering the retrieved list of entity objects by removing any entity object representing a linked concept expressly related to the user node within the social graph;

(iii) referencing a plurality of sub-concepts of remaining entity objects within the filtered list of entity objects by accessing linked entity objects within the entity graph representing the plurality of sub-concepts, the linked entity objects expressly related to the remaining entity objects; and

(iv) querying the remaining entity objects within the filtered list of entity objects and the plurality of sub-concepts of the remaining entity objects for metadata of the respective entity objects.

3. The method of claim 1 wherein retrieving socially relevant data for the user within the social graph of the social network system further includes accessing one or more linked concept nodes within the entity graph expressly related to the user node within the social graph.

4. The method of claim 1, wherein retrieving socially relevant data for the user node of the user within the social graph, comprises at least:

querying for the socially relevant data of the user node describing relationships to friend nodes of the user node within the social graph, each of the friend nodes being another user node within the social graph of the social network system which is expressly linked to the user node of the user;

querying for one or more linked concept nodes within the entity graph expressly related to the user node within the social graph according to an express relationship between a user object representing the user node within the social graph and an entity object within the entity graph representing the one or more linked concept nodes within the entity graph; and

querying objects representing the friend nodes expressly linked to the user node for one or more friend linked concept nodes within the entity graph, each expressly linked to the respective friend nodes and not linked to the user node.

5. The method of claim 1, wherein identifying intersects between the plurality of concepts within the entity graph and the socially relevant data for the user node within the social graph, comprises:

querying entity objects representing the plurality of concepts for data in common with the socially relevant data retrieved for the user, wherein the data in common between the plurality of concepts and the socially relevant data retrieved for the user comprises any data element existing in two or more of;

(a) the entity objects representing the plurality of concepts which are not expressly related to the user node;

(b) friend objects representing the friend nodes of the user node and expressly related to the user node;

or

(c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node.

6. The method of claim 5, wherein querying the entity objects representing the plurality of concepts for the data in common with the socially relevant data retrieved for the user node, comprises:

creating a joined table from;

(a) the entity objects representing the plurality of concepts which are not expressly linked to the user node, (b) the friend objects representing the friend nodes of the user node and expressly linked to the user node, and (c) the linked entity objects representing the one or more linked concept nodes expressly related to the user object representing the user node; and

searching the joined table for redundant fields indicating the data in common between at least two or more of (a) the entity objects representing the plurality of concepts which are not expressly linked to the user node, (b) the friend objects representing the friend nodes of the user node and expressly linked to the user node, and (c) the linked entity objects representing the one or more linked concept nodes expressly related to the user object representing the user node.

7. The method of claim 1, further comprising:

generating one or more sub-concept search boxes associated with one or more of the selected sub-concept recommendations; and

dynamically generating, for display on the user interface associated with the user in response to the search request, the concept page view with the selected sub-concept recommendations and the one or more sub-concept search boxes for the plurality of concepts within the entity graph.

8. The method of claim 1, further comprising:

identifying a socially relevant, real-world event, from the socially relevant data, associated with one or more of the friend nodes and one or more of the selected sub-concept recommendations, wherein the socially relevant, real-world event comprises a physical event, by a friend, related to the one or more of the sub-concept recommendations; and

dynamically generating, for display on the user interface associated with the user in response to the search request, the concept page view with the selected sub-concept recommendations and the identified socially relevant, real-world event for the plurality of concepts within the entity graph.

9. The method of claim 1, wherein the search request includes:

receiving a page load request from a search engine external to the social network system, the page load request specifying the search parameters and a link corresponding to the page load request;

identifying a user associated with the page load request based on cookie or session data at a client device corresponding to the page load request from the search engine external to the social network system by identifying the user node based on the identified user associated with the page load request, wherein the user associated with the page load out request is a known user of the social network system having existing authentication active for the user associated with the page load request at the client device corresponding to the page load out;

retrieving the socially relevant data for the user node within the social graph by identifying the user node based on the identified user associated with the page load request; and

wherein displaying the sub-concept recommendations for the plurality of concepts comprises dynamically generating the concept page view at the link corresponding to the page load request within the user interface.

10. The method of claim 9, further comprising:

identifying a user associated with the page load request based on cookie or session data at a client device corresponding to the page load request from the search engine external to the social network system;

wherein the user associated with the page load request is one of (a) a known user of the social network system lacking existing authentication for the user at the client device corresponding to the page load request or (b) an unknown user of the social network system;

retrieving the socially relevant data for a user'"'"'s node within the social graph by accessing a temporary unauthenticated user node for the user associated with the page load request; and

providing the display of the sub-concept recommendations for the plurality of concepts to the user associated with the page load request and the temporary unauthenticated user node of the social network system at the user display of the client device corresponding to the page load request from the search engine external to the social network system.

11. A non-transitory computer readable storage medium having instructions stored thereon that, when executed by a processor of a social network system, the instructions cause the social network system to perform operations comprising:

receiving, from a user of the social network system, a search request including one or more search parameters;

retrieving contextually relevant data for a plurality of concepts within an entity graph of the social network system based upon the one or more search parameters, wherein the plurality of concepts comprises two or more of places, people, events, and monetization recommendations;

retrieving socially relevant data for the user within a social graph of the social network system by accessing a node for the user in the social graph that includes socially relevant data describing relationships to friend nodes of the user within the social graph and by accessing one or more concept nodes within the entity graph expressly related to the user node within the social graph or to the friend nodes of the user or both;

identifying intersects between the plurality of concepts within the entity graph and the socially relevant data for the user node within the social graph;

selecting sub-concept recommendations for the plurality of concepts within the entity graph based on the intersects identified by;

creating a common dataset from;

(a) entity objects representing the sub-concept recommendations for the plurality of concepts which are not expressly related to the user node, (b) friend objects representing the friend nodes of the user and expressly related to the user node, and (c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node;

scoring entries in the common dataset by applying a TFIDF (Term Frequency-Inverse Document Frequency) utility to the common dataset; and

selecting the sub-concept recommendations for the plurality of concepts based on the scores of the entries, wherein the sub-concept recommendations comprise of instances of the plurality of concepts; and

dynamically generating, for display on a user interface associated with the user in response to the search request, a concept page view with the selected sub-concept recommendations for the plurality of concepts within the entity graph.

12. The non-transitory computer readable storage medium of claim 11, wherein retrieving socially relevant data for the user node of the user within the social graph of the social network system further includes accessing one or more linked concept nodes within the entity graph expressly related to the user node within the social graph.

13. The non-transitory computer readable storage medium of claim 11, the instructions, when executed by the processor of the social network system, cause the social network system to perform operations comprising:

querying entity objects representing the plurality of concepts for data in common with the socially relevant data retrieved for the user, wherein the data in common between the plurality of concepts and the socially relevant data retrieved for the user comprises any data element existing in two or more of;

(a) the entity objects representing the plurality of concepts which are not expressly related to the user node;

(b) friend objects representing the friend nodes of the user node and expressly related to the user node;

or

(c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node.

14. The non-transitory computer readable storage medium of claim 11, wherein selecting sub-concept recommendations for the plurality of concepts within the entity graph based on the intersects identified, comprises:

ranking the sub-concept recommendations for the plurality of concepts based on at least one of;

global reference counts for each of the sub-concept recommendations for the plurality of concepts;

global trending statistics for each of the sub-concept recommendations for the plurality of concepts; and

tracking event data for the sub-concept recommendations, wherein the tracked event data comprises one or more of page loads, link clicks, and mouse focus events in relation to the sub-concept recommendations; and

selecting the sub-concept recommendations for the plurality of concepts based on the rankings of the sub-concept recommendations.

15. The non-transitory computer readable storage medium of claim 11, the instructions, when executed by the processor of the social network system, cause the social network system to perform operations comprising:

identifying a socially relevant, real-world event, from the socially relevant data, associated with one or more of the friend nodes and one or more of the selected sub-concept recommendations, wherein the socially relevant, real-world event comprises a physical event, by a friend, related to the one or more of the sub-concept recommendations; and

dynamically generating, for display on the user interface associated with the user in response to the search request, the concept page view with the selected sub-concept recommendations and the identified socially relevant, real-world event for the plurality of concepts within the entity graph.

16. The non-transitory computer readable storage medium of claim 11, wherein selecting the sub-concept recommendations for the plurality of concepts within the entity graph based on the intersects identified, comprises:

scoring each of the sub-concept recommendations for the plurality of concepts according to frequency of commonality with the socially relevant data retrieved for the user;

identifying preferred contextual relationship types for each of the sub-concept recommendations for the plurality of concepts from a pre-determined list;

retrieving weighting values for the preferred contextual relationship types;

proportionally increasing the scoring of each of the sub-concept recommendations for the plurality of concepts according to the retrieved weighting values for those of the sub-concept recommendations for the plurality of concepts matching the preferred contextual relationship types; and

selecting the sub-concept recommendations for the plurality of concepts based on the scores of each of the sub-concept recommendations.

17. The non-transitory computer readable storage medium of claim 11, the instructions, when executed by the processor of the social network system, cause the social network system to perform operations comprising:

dynamically generating a global concept page view comprising global sub-concept recommendations comprising a confidence value corresponding to the sub-concept recommendations; and

adjusting the confidence value corresponding to the global sub-concept recommendations by;

increasing the confidence value of the global sub-concept recommendations upon receiving affirmative feedback for the global sub-concept recommendations from the global concept page view; and

decreasing the confidence value of the global sub-concept recommendations upon receiving negative feedback for the global sub-concept recommendations from the global concept page view; and

selecting the sub-concept recommendations for the plurality of concepts within the entity graph based on the confidence value of the corresponding global sub-concept recommendations.

18. A system, comprising:

a processing device; and

a memory coupled to the processing device, the memory storing instructions which, when executed by the processing device, causes the system to;

receive, from a user of a social network system, a search request including one or more search parameters;

retrieve contextually relevant data for a plurality of concepts within an entity graph of the social network system based upon the one or more search parameters, wherein the plurality of concepts comprises two or more of places, people, events, and monetization recommendations;

retrieve socially relevant data for the user within a social graph of the social network system by accessing a node for the user in the social graph that includes socially relevant data describing relationships to friend nodes of the user within the social graph and by accessing one or more concept nodes within the entity graph expressly related to the user node within the social graph or to the friend nodes of the user or both;

identify intersects between the plurality of concepts within the entity graph and the socially relevant data for the user node within the social graph;

select sub-concept recommendations for the plurality of concepts within the entity graph based on the intersects identified by;

creating a common dataset from;

(a) entity objects representing the sub-concept recommendations for the plurality of concepts which are not expressly related to the user node, (b) friend objects representing the friend nodes of the user and expressly related to the user node, and (c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node;

scoring entries in the common dataset by applying a TF-IDF (Term Frequency-Inverse Document Frequency) utility to the common dataset; and

selecting the sub-concept recommendations for the plurality of concepts based on the scores of the entries, wherein the sub-concept recommendations comprise of instances of the plurality of concepts; and

dynamically generating, for display on a user interface associated with the user in response to the search request, a concept page view with the selected sub-concept recommendations for the plurality of concepts within the entity graph.

19. The system of claim 18, wherein the instructions, when executed by the processing device, cause the system to retrieve socially relevant data for the user within the social graph of the social network system further includes accessing one or more linked concept nodes within the entity graph expressly related to the user node within the social graph.

20. The system of claim 18, wherein the instructions, when executed by the processing device, cause the system to identify intersects between the plurality of concepts within the entity graph and the socially relevant data for the user node within the social graph by:

querying entity objects representing the plurality of concepts for data in common with the socially relevant data retrieved for the user, wherein the data in common between the plurality of concepts and the socially relevant data retrieved for the user node comprises any data element existing in two or more of;

(a) the entity objects representing the plurality of concepts which are not expressly related to the user node;

(b) friend objects representing the friend nodes of the user node and expressly related to the user node;

or

(c) linked entity objects representing one or more linked concept nodes expressly related to a user object representing the user node.