Creating ontologies by analyzing natural language texts

• US 9,626,358 B2
• Filed: 01/02/2015
• Issued: 04/18/2017
• Est. Priority Date: 11/26/2014
• Status: Active Grant

• Current Assignee: ABBYY Development LLC
• Original Assignee: ABBYY InfoPoisk LLC
• Inventors: Danielyan, Tatiana
• Primary Examiner(s): Pham, Thierry L
• Application Number:
  US14/588,644
• Publication Number:
  US 20160147736A1
• Time in Patent Office: 837 Days
• Field of Search: 704 8, 704 9, 704257
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/414 Extracting the geometrical structure, e.g. layout tree; Block segmentation, e.g. bounding boxes for graphics or text
• CPC Class Codes: G06V 30/418 Document matching, e.g. of document images

Abstract

Systems and methods for creating ontologies by analyzing natural language texts. An example method comprises: receiving a plurality of semantic structures associated with a text corpus; identifying a first semantic structure and a second semantic structure, wherein the first semantic structure comprises a first substructure and a second substructure, wherein the second semantic structure comprises a third substructure and a fourth substructure, and wherein the first substructure is similar to the third substructure in view of a first similarity criterion; and responsive to determining that the second substructure is similar to the fourth substructure in view of a second similarity criterion, associating, with a certain concept of an ontology associated with the text corpus, objects represented by the second substructure and the fourth substructure.

Claims

1. A method, comprising:

receiving a plurality of semantic structures associated with a text corpus;

identifying, by a processing device, a first semantic structure and a second semantic structure, wherein the first semantic structure comprises a first substructure and a second substructure, wherein the second semantic structure comprises a third substructure and a fourth substructure, and wherein the first substructure is similar to the third substructure in view of a first similarity criterion; and

responsive to determining that the second substructure is similar to the fourth substructure in view of a second similarity criterion, associating, with a certain concept of an ontology associated with the text corpus, objects represented by the second substructure and the fourth substructure.

2. The method of claim 1, wherein the ontology comprises one or more concepts, each concept associated with one or more instances of the concept represented by one or more objects.

3. The method of claim 1, wherein the first substructure comprises a left context and a right context surrounding the second substructure.

4. The method of claim 1, wherein the second substructure comprises a left context and a right context surrounding the fourth substructure.

5. The method of claim 1, wherein determining that the second substructure is similar to the fourth substructure comprises:

identifying a third semantic structure and a fourth semantic structure, wherein the third semantic structure comprises the second substructure and a fifth substructure, wherein the fourth semantic structure comprises the fourth substructure and a sixth substructure, and wherein the firth substructure is similar to the sixth substructure in view of the first similarity criterion.

6. The method of claim 1, wherein at least one of the first semantic structure and the second semantic structure is represented by a graph comprising a plurality of nodes corresponding to a plurality of semantic classes and a plurality of edges corresponding to a plurality of semantic relationships.

7. The method of claim 1, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantic classes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantic classes associated with a second plurality of nodes of a second graph representing the second semantic structure.

8. The method of claim 1, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantemes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantemes associated with a second plurality of nodes of a second graph representing the second semantic structure.

9. The method of claim 1, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of deep slots associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of deep slots associated with a second plurality of nodes of a second graph representing the second semantic structure.

10. The method of claim 1, further comprising producing the plurality of semantic structures by performing a syntactico-semantic analysis of the text corpus.

11. A system, comprising:

a memory;

a processor, coupled to the memory, the processor configured to;

receiving a plurality of semantic structures associated with a text corpus;

identify a first semantic structure and a second semantic structure, wherein the first semantic structure comprises a first substructure and a second substructure, wherein the second semantic structure comprises a third substructure and a fourth substructure, and wherein the first substructure is similar to the third substructure in view of a first similarity criterion; and

responsive to determining that the second substructure is similar to the fourth substructure in view of a second similarity criterion, associate, with a certain concept of an ontology associated with the text corpus, objects represented by the second substructure and the fourth substructure.

12. The system of claim 11, wherein determining that the second substructure is similar to the fourth substructure comprises:

identifying a third semantic structure and a fourth semantic structure, wherein the third semantic structure comprises the second substructure and a fifth substructure, wherein the fourth semantic structure comprises the fourth substructure and a sixth substructure, and wherein the firth substructure is similar to the sixth substructure in view of the first similarity criterion.

13. The system of claim 11, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantic classes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantic classes associated with a second plurality of nodes of a second graph representing the second semantic structure.

14. The system of claim 11, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantemes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantemes associated with a second plurality of nodes of a second graph representing the second semantic structure.

15. The system of claim 11, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of deep slots associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of deep slots associated with a second plurality of nodes of a second graph representing the second semantic structure.

16. A computer-readable non-transitory storage medium comprising executable instructions that, when executed by a computing device, cause the computing device to perform operations comprising:

receiving a plurality of semantic structures associated with a text corpus;

identifying, by a processing device, a first semantic structure and a second semantic structure, wherein the first semantic structure comprises a first substructure and a second substructure, wherein the second semantic structure comprises a third substructure and a fourth substructure, and wherein the first substructure is similar to the third substructure in view of a first similarity criterion; and

responsive to determining that the second substructure is similar to the fourth substructure in view of a second similarity criterion, associating, with a certain concept of an ontology associated with the text corpus, objects represented by the second substructure and the fourth substructure.

17. The computer-readable non-transitory storage medium of claim 16, wherein determining that the second substructure is similar to the fourth substructure comprises:

identifying a third semantic structure and a fourth semantic structure, wherein the third semantic structure comprises the second substructure and a fifth substructure, wherein the fourth semantic structure comprises the fourth substructure and a sixth substructure, and wherein the firth substructure is similar to the sixth substructure in view of the first similarity criterion.

18. The computer-readable non-transitory storage medium of claim 16, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantic classes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantic classes associated with a second plurality of nodes of a second graph representing the second semantic structure.

19. The computer-readable non-transitory storage medium of claim 16, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of semantemes associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of semantemes associated with a second plurality of nodes of a second graph representing the second semantic structure.

20. The computer-readable non-transitory storage medium of claim 16, wherein identifying the first semantic structure and the second semantic structure comprises comparing a first plurality of deep slots associated with a first plurality of nodes of a first graph representing the first semantic structure to a second plurality of deep slots associated with a second plurality of nodes of a second graph representing the second semantic structure.