Universal concept graph for a social networking service

• US 10,380,145 B2
• Filed: 02/24/2016
• Issued: 08/13/2019
• Est. Priority Date: 02/24/2016
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Kenthapadi, Krishnaram, Borisyuk, Fedor Vladimirovich, Jain, Parul
• Primary Examiner(s): Kuddus, Daniel A
• Application Number:
  US15/052,359
• Publication Number:
  US 20170242909A1
• Time in Patent Office: 1,266 Days
• CPC Class Codes: G06F 16/287 Visualization; Browsing
• CPC Class Codes: G06F 16/9558 Details of hyperlinks; Management of linked annotations
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 67/306 User profiles

Abstract

A machine may be configured to generate a universal concept graph. For example, the machine generates an internal set of concept phrases based on an internal dataset including content from internal documents associated with a social networking service (SNS). The machine generates an external set of concept phrases based on an external dataset including content from external documents that are external to the SNS. The machine generates a set of nodes for a universal concept graph based on performing a union operation of the internal set of concept phrases and the external set of concept phrases. Each node corresponds to a particular concept phrase. The machine generates a set of edges among a plurality of nodes of the set of nodes based on one or more relationship indicators for pairs of nodes. The machine generates the universal concept graph based on the set of nodes and the set of edges.

Claims

1. A method comprising:

generating an internal set of concept phrases based on an internal dataset that includes content from one or more internal documents associated with a social networking service (SNS), the concept phrases included in the internal set of concept phrases being one or more terms extracted from the one or more internal documents;

generating an external set of concept phrases based on an external dataset that includes content from one or more external documents, the one or more external documents accessed from one or more sources that are external to the SNS, the concept phrases included in the external set of concept phrases being one or more terms extracted from the one or more external documents, the generating of the external set of concept phrases including;

identifying one or more titles of the one or more external documents included in the external dataset, and

generating one or more tokens for each of the one or more titles, the generating of the one or more tokens comprising;

parsing the one or more titles of the one or more external documents, extracting the one or more terms from the one or more parsed titles of the one or more external documents, and identifying the one or more terms as the one or more tokens;

generating a set of nodes for a universal concept graph based on performing a union operation of the internal set of concept phrases and the external set of concept phrases, each node corresponding to a particular concept phrase;

generating a set of edges among a plurality of nodes of the set of nodes based on one or more relationship indicators for pairs of nodes of the set of nodes;

generating, using one or more hardware processors, the universal concept graph based on the set of nodes and the set of edges among the plurality of nodes; and

causing presentation of at least a portion of the universal concept graph in a user interface of a device associated with a user, the causing presentation of the at least the portion of the universal concept graph including causing display of one or more user interface elements for receiving input data pertaining to one or more nodes or one or more edges of the universal concept graph, a receiving of the input data via the one or more user interface elements causing generation and association of metadata with the one or more nodes or the one or more edges of the universal concept graph based on the input data received via the one or more user interface elements.

2. The method of claim 1, wherein the particular concept phrase corresponds to an identifier of an internal document or an identifier of an external document.

3. The method of claim 1, wherein the generating of the internal set of concept phrases includes:

generating a first set of tokens based on a content of an internal document of the one or more documents, each token in the first set of tokens including one or more keywords comprised in the internal document; and

generating a canonical version of one or more tokens in the first set of tokens based on mapping the one or more tokens to one or more external concept phrases in the external dataset,

wherein the internal set of concept phrases includes the canonical version of the one or more tokens.

4. The method of claim 3, wherein the internal document is a member profile of a member of the SNS.

5. The method of claim 1, wherein the internal set of concept phrases includes at least one of a skill, an occupation title, an educational course name, a location, a company name, or a university name.

6. The method of claim 1, wherein the external set of concept phrases includes at least one of an article title of an article included in the external dataset or a keyword identified by a hyperlink in the article.

7. The method of claim 1, further comprising:

determining a relationship indicator of the one or more relationship indicators based on analyzing contents of documents associated with a pair of nodes.

8. The method of claim 7, wherein the determining of the relationship indicator comprises:

identifying a hyperlink included in a first document associated with a first document title that corresponds to a first node of the pair of nodes, the hyperlink being directed to a second document associated with a second document title that corresponds to a second node of the pair of nodes.

9. The method of claim 7, wherein the determining of the relationship indicator includes:

determining a similarity value associated with the pair of nodes based on a content similarity between a first document associated with a first document title that corresponds to a first node of the pair of nodes and a second document associated with a second document title that corresponds to a second node of the pair of nodes.

10. The method of claim 1, further comprising:

determining a percentage of content overlap between a first document associated with a first title that corresponds to a first node in a pair of nodes of the set of nodes and a second document associated with a second title that corresponds to a second node in the pair of nodes, the pair of nodes being connected by an edge of the set of edges; and

associating an edge weight with the edge based on the percentage of content overlap between the first document and the second document.

11. The method of claim 1, further comprising:

determining a percentage of members of the SNS associated with member profiles that include a first concept phrase corresponding to a first node of a pair of nodes of the set of nodes and a second concept phrase corresponding to a second node of the pair of nodes, the pair of nodes being connected by an edge of the set of edges; and

associating an edge weight with the edge based on the percentage of members of the SNS associated with member profiles that include the first concept phrase and the second concept phrase.

12. A system comprising:

one or more hardware processors; and

a non-transitory machine-readable medium for storing instructions that, when executed by the one or more hardware processors, cause the system to perform operations comprising;

generating an internal set of concept phrases based on an internal dataset that includes content from one or more internal documents associated with a social networking service (SNS), the concept phrases included in the internal set of concept phrases being one or more terms extracted from the one or more internal documents;

generating an external set of concept phrases based on an external dataset that includes content from one or more external documents, the one or more external documents accessed from one or more sources that are external to the SNS, the concept phrases included in the external set of concept phrases being one or more terms extracted from the one or more external documents, the generating of the external set of concept phrases including;

identifying one or more titles of the one or more external documents included in the external dataset, and

generating one or more tokens for each of the one or more titles, the generating of the one or more tokens comprising;

parsing the one or more titles of the one or more external documents, extracting the one or more terms from the one or more parsed titles of the one or more external documents, and identifying the one or more terms as the one or more tokens;

generating a set of nodes for a universal concept graph based on performing a union operation of the internal set of concept phrases and the external set of concept phrases, each node corresponding to a particular concept phrase;

generating a set of edges among a plurality of nodes of the set of nodes based on one or more relationship indicators for pairs of nodes of the set of nodes;

generating the universal concept graph based on the set of nodes and the set of edges among the plurality of nodes; and

causing presentation of at least a portion of the universal concept graph in a user interface of a device associated with a user, the causing presentation of the at least the portion of the universal concept graph including causing display of one or more user interface elements for receiving input data pertaining to one or more nodes or one or more edges of the universal concept graph, a receiving of the input data via the one or more user interface elements causing generation and association of metadata with the one or more nodes or the one or more edges of the universal concept graph based on the input data received via the one or more user interface elements.

13. The system of claim 12, wherein the generating of the internal set of concept phrases includes:

generating a first set of tokens based on a content of an internal document of the one or more documents, each token in the first set of tokens including one or more keywords comprised in the internal document; and

generating a canonical version of one or more tokens in the first set of tokens based on mapping the one or more tokens to one or more external concept phrases in the external dataset,

wherein the internal set of concept phrases includes the canonical version of the one or more tokens.

14. The system of claim 12, wherein the operations further comprise:

determining a relationship indicator of the one or more relationship indicators based on analyzing contents of documents associated with a pair of nodes.

15. The system of claim 14, wherein the determining of the relationship indicator comprises:

identifying a hyperlink included in a first document associated with a first document title that corresponds to a first node of the pair of nodes, the hyperlink being directed to a second document associated with a second document title that corresponds to a second node of the pair of nodes.

16. The system of claim 12, wherein the operations further comprise:

determining a percentage of content overlap between a first document associated with a first title that corresponds to a first node in a pair of nodes of the set of nodes and a second document associated with a second title that corresponds to a second node in the pair of nodes, the pair of nodes being connected by an edge of the set of edges; and

associating an edge weight with the edge based on the percentage of content overlap between the first document and the second document.

17. A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:

generating an internal set of concept phrases based on an internal dataset that includes content from one or more internal documents associated with a social networking service (SNS), the concept phrases included in the internal set of concept phrases being one or more terms extracted from the one or more internal documents;

generating an external set of concept phrases based on an external dataset that includes content from one or more external documents, the one or more external documents accessed from one or more sources that are external to the SNS, the concept phrases included in the external set of concept phrases being one or more terms extracted from the one or more external documents, the generating of the external set of concept phrases including;

identifying one or more titles of the one or more external documents included in the external dataset and

generating one or more tokens for each of the one or more titles, the generating of the one or more tokens comprising;

parsing the one or more titles of the one or more external documents, extracting the one or more terms from the one or more parsed titles of the one or more external documents, and identifying the one or more terms as the one or more tokens;

generating a set of nodes for a universal concept graph based on performing a union operation of the internal set of concept phrases and the external set of concept phrases, each node corresponding to a particular concept phrase;

generating a set of edges among a plurality of nodes of the set of nodes based on one or more relationship indicators for pairs of nodes of the set of nodes;

generating the universal concept graph based on the set of nodes and the set of edges among the plurality of nodes; and

causing presentation of at least a portion of the universal concept graph in a user interface of a device associated with a user, the causing presentation of the at least the portion of the universal concept graph including causing display of one or more user interface elements for receiving input data pertaining to one or more nodes or one or more edges of the universal concept graph, a receiving of the input data via the one or more user interface elements causing generation and association of metadata with the one or more nodes or the one or more edges of the universal concept graph based on the input data received via the one or more user interface elements.