Determining an object referenced within informal online communications

Patent metadata

Field	Value
Publication number	US-10296527-B2
Application number	US-201514962232-A
Country	US
Kind code	B2
Filing date	Dec 8, 2015
Priority date	Dec 8, 2015
Publication date	May 21, 2019
Grant date	May 21, 2019

Abstract

A method for determining an object referenced within a set of one or more informal online communications comprises: generating a knowledge graph for a company based at least on formal online communications, the knowledge graph comprising a plurality of node elements, and the knowledge graph further comprising, for each node element of the knowledge graph, a corresponding halo comprising one or more words which are temporally proximate to that node element within the formal online communications; for each node element of the knowledge graph which is determined to be present in a given informal online communication, detecting a halo comprising one or more words which are temporally proximate to that node element within the given informal online communication; and identifying which of the plurality of node elements has a corresponding halo within the knowledge graph most similar to the detected halo, wherein the identified node element is the referenced object.

First claim

What is claimed is:

1. A method for determining an object referenced within a set of one or more informal online communications, the method comprising:
generating a knowledge graph for at least a given company based at least on a set of one or more formal online communications comprising at least one of one or more press releases about the given company and one or more news items about the given company, the knowledge graph comprising a plurality of node elements and relationships extracted from an enterprise resource planning database for the given company, and the knowledge graph further comprising, for each node element of the knowledge graph, a corresponding halo comprising one or more words which are temporally proximate to that node element within at least one of the one or more formal online communications;
determining whether at least one of the node elements of at least the knowledge graph for the given company is present in a given informal online communication within the set of one or more informal online communications;
for each node element of the knowledge graph for the given company which is determined to be present in the given informal online communication, detecting a halo comprising one or more words which are temporally proximate to that node element within the given informal online communication; and
identifying which of the plurality of node elements has the corresponding halo, within the knowledge graph for the given company, most similar to the detected halo, wherein the identified node element is determined to be the referenced object;
wherein generating a knowledge graph for at least the given company comprises:
selecting the plurality of node elements from the enterprise resource planning database for the given company;
calculating whether each pair within the plurality of node elements has a relationship within the enterprise resource planning database for the given company; and
generating the corresponding halo for each node element by extracting one or more keywords from any formal online communication within the set of one or more formal online communications which includes temporally proximate references to that node element and to the given company.

2. The method of claim 1 , wherein generating a knowledge graph for at least the given company comprises generating a plurality of knowledge graphs, each of the plurality of knowledge graphs corresponding to a respective at least one of a plurality of companies.

3. The method of claim 2 , further comprising repeating the determining, detecting and identifying steps for each of the plurality of knowledge graphs.

4. The method of claim 1 , wherein each corresponding halo within the knowledge graph represents a temporal event profile.

5. The method of claim 1 , further comprising performing keyword detection on the given informal communication utilizing at least one of context and topic segmentation.

6. The method of claim 1 , wherein identifying which of the plurality of node elements has the corresponding halo most similar to the detected halo comprises computing an overlap percentage between the detected halo and at least one of the plurality of node elements.

7. The method of claim 1 , wherein identifying which of the plurality of node elements has the corresponding halo most similar to the detected halo further comprises weighting the overlap percentage based at least in part on a temporal proximity between the given informal communication and at least one formal online communication, within the set of one or more formal online communications, used to generate the corresponding halo.

8. The method of claim 1 , further comprising, when at least one of the node elements of the knowledge graph is determined to be present in the given informal online communication, performing word segmentation on the given informal online communication.

9. The method of claim 1 , wherein the plurality of node elements from the enterprise resource planning database comprise at least one of products, fields, and industries for the given company.

10. The method of claim 1 , wherein the given informal communication comprises at least one of a chat, message board, and social media.

11. An apparatus for determining an object referenced within a set of one or more informal online communications, the apparatus comprising:
a memory; and
at least one processor coupled with the memory and operative:
to generate a knowledge graph for at least a given company based at least on a set of one or more formal online communications comprising at least one of one or more press releases about the given company and one or more news items about the given company, the knowledge graph comprising a plurality of node elements and relationships extracted from an enterprise resource planning database for the given company, and the knowledge graph further comprising, for each node element of the knowledge graph, a corresponding halo comprising one or more words which are temporally proximate to that node element within at least one of the one or more formal online communications;
to determine whether at least one of the node elements of at least the knowledge graph for the given company is present in a given informal online communication within the set of one or more informal online communications;
for each node element of the knowledge graph for the given company which is determined to be present in the given informal online communication, to detect a halo comprising one or more words which are temporally proximate to that node element within the given informal online communication; and
to identify which of the plurality of node elements has the corresponding halo within the knowledge graph for the given company most similar to the detected halo, wherein the identified node element is determined to be the referenced object;
wherein generating a knowledge graph for at least the given company comprises:
selecting the plurality of node elements from the enterprise resource planning database for the given company;
calculating whether each pair within the plurality of node elements has a relationship within the enterprise resource planning database for the given company; and
generating the corresponding halo for each node element by extracting one or more keywords from any formal online communication within the set of one or more formal online communications which includes temporally proximate references to that node element and to the given company.

12. The apparatus of claim 11 , wherein generating a knowledge graph for at least the given company comprises generating a plurality of knowledge graphs, each of the plurality of knowledge graphs corresponding to a respective at least one of a plurality of companies.

13. The apparatus of claim 12 , wherein the at least one processor is further operative to repeat the determining, detecting and identifying steps for each of the plurality of knowledge graphs.

14. The apparatus of claim 11 , wherein the plurality of node elements from the enterprise resource planning database comprise at least one of products, fields, and industries for the given company.

15. The apparatus of claim 11 , wherein the given informal communication comprises at least one of a chat, message board, and social media.

16. A computer program product comprising a non-transitory machine-readable storage medium having machine-readable program code embodied therewith, said machine-readable program code comprising:
machine-readable program code configured:
to generate a knowledge graph for at least a given company based at least on a set of one or

Assignees

• IBM

Inventors

• Chu Stephen M
• Duan Ning
• Gong Min
• Qiu Yun Jie
• Yan Junchi

Classifications

• G06F17/30705

  Physics · mapped topic

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title

• G06F16/33 Primary

  Querying · CPC title

• G06F16/383

  using metadata automatically derived from the content · CPC title

• G06F16/35

  Clustering; Classification · CPC title