System, method and computer program for using a multi-tiered knowledge representation model

• US 7,844,565 B2
• Filed: 06/04/2009
• Issued: 11/30/2010
• Est. Priority Date: 03/30/2005
• Status: Active Grant

• Current Assignee: Primal Fusion Inc.
• Original Assignee: Primal Fusion Inc.
• Inventors: Sweeney, Peter
• Primary Examiner(s): HOLMES, MICHAEL B
• Application Number:
  US12/477,994
• Publication Number:
  US 20090327205A1
• Time in Patent Office: 544 Days
• Field of Search: 706/54
• US Class Current: 706/54
• CPC Class Codes: G06F 16/84 Mapping; Conversion
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

A method (system and computer program product) performs facet classification synthesis to relate concepts represented by concept definitions defined in accordance with a faceted data set comprising facets, facet attributes, and facet attributes hierarchies. Dimensional concept relationships are expressed between the concept definitions. Two concept definitions are determined to be related in a particular dimensional concept relationship by examining whether at least one of explicit relationships and implicit relationships exist in the faceted data set between the respective facet attributes of the two concept definitions.

Claims

1. A computer implemented method for generating a plurality of concept definitions through multiple tiers of abstraction using a domain of information, wherein the domain of information is a faceted domain that comprises a plurality of facets each having facet attributes, the method comprising:

extracting a plurality of concepts from the domain of information;

for each concept of the plurality of concepts, extracting at least one atomic concept associated with the concept from the domain of information, to extract a plurality of atomic concepts;

for each atomic concept of the plurality of atomic concepts, extracting at least one facet attribute associated with the atomic concept, to extract a plurality of facet attributes; and

generating a plurality of concept definitions, each of the plurality of concept definitions corresponding to a concept of the plurality of concepts, in accordance with a multi-tiered knowledge representation model, the multi-tiered knowledge representation model comprising at least three tiers, wherein a first tier of the multi-tiered knowledge representation model corresponds to the concepts, a second tier of the multi-tiered knowledge representation model corresponds to the atomic concepts, and a third tier of the multi-tiered knowledge representation model corresponds to the facet attributes, the generating comprising including at least one of the plurality of facet attributes in each of the concept definitions, in accordance with the associations between the extracted concepts, atomic concepts and facet attributes.

2. The method of claim 1, wherein the atomic concepts correspond to keywords, and the facet attributes correspond to morphemes.

3. The method of claim 1, further comprising inferring facet attribute relationships from the plurality of concept definitions and concept relationships obtained from the domain of information.

4. The method of claim 1, further comprising revising at least some of the plurality of concept definitions based at least in part on relationships between facet attributes of the at least some of the plurality of concept definitions through the multi-tiered knowledge representation model.

5. The method of claim 1, further comprising:

determining whether any explicit relationships exist between the plurality of concept definitions, wherein an explicit relationship is determined to exist between any two of the plurality of concept definitions if the two of the plurality of concept definitions share at least one common facet attribute or each has a facet attribute of the same lineage in at least one facet attribute hierarchy of the plurality of facet attributes;

determining whether any implicit relationships exist between the plurality of concept definitions, wherein an implicit relationship between any two of the plurality of concept definitions is determined to exist based on a statistical identification of a relationship between a facet attribute in a first of the two of the plurality of concept definitions and a facet attribute in a second of the two of the plurality of concept definitions; and

when it is determined that at least one explicit relationship and/or at least one implicit relationship exists between any two of the plurality of concept definitions, synthesizing a relationship between the two of the plurality of concept definitions.

6. The method of claim 5, further comprising creating a dimensional concept hierarchy, based at least in part on the synthesized relationships between concept definitions.

7. The method of claim 6, further comprising classifying content of the domain of information based on the dimensional concept hierarchy.

8. A computer system comprising:

at least one memory that stores processor-executable instructions for generating a plurality of concept definitions through multiple tiers of abstraction using a domain of information, wherein the domain of information is a faceted domain that comprises a plurality of facets each having facet attributes; and

at least one hardware processor, operatively coupled to the at least one memory, that executes the instructions to;

extract a plurality of concepts from the domain of information;

for each concept of the plurality of concepts, extract at least one atomic concept associated with the concept from the domain of information, to extract a plurality of atomic concepts;

for each atomic concept of the plurality of atomic concepts, extract at least one facet attribute associated with the atomic concept, to extract a plurality of facet attributes; and

generate a plurality of concept definitions, each of the plurality of concept definitions corresponding to a concept of the plurality of concepts, in accordance with a multi-tiered knowledge representation model, the multi-tiered knowledge representation model comprising at least three tiers, wherein a first tier of the multi-tiered knowledge representation model corresponds to the concepts, a second tier of the multi-tiered knowledge representation model corresponds to the atomic concepts, and a third tier of the multi-tiered knowledge representation model corresponds to the facet attributes, the generating comprising including at least one of the plurality of facet attributes in each of the concept definitions, in accordance with the associations between the extracted concepts, atomic concepts and facet attributes.

9. The computer system of claim 8, wherein the atomic concepts correspond to keywords, and the facet attributes correspond to morphemes.

10. The computer system of claim 8, wherein the at least one processor further executes the instructions to infer facet attribute relationships from the plurality of concept definitions and concept relationships obtained from the domain of information.

11. The computer system of claim 8, wherein the at least one processor further executes the instructions to revise at least some of the plurality of concept definitions based at least in part on relationships between facet attributes of the at least some of the plurality of concept definitions through the multi-tiered knowledge representation model.

12. The computer system of claim 8, wherein the at least one processor further executes the instructions to:

determine whether any explicit relationships exist between the plurality of concept definitions, wherein an explicit relationship is determined to exist between any two of the plurality of concept definitions if the two of the plurality of concept definitions share at least one common facet attribute or each has a facet attribute of the same lineage in at least one facet attribute hierarchy of the plurality of facet attributes;

determine whether any implicit relationships exist between the plurality of concept definitions, wherein an implicit relationship between any two of the plurality of concept definitions is determined to exist based on a statistical identification of a relationship between a facet attribute in a first of the two of the plurality of concept definitions and a facet attribute in a second of the two of the plurality of concept definitions; and

when it is determined that at least one explicit relationship and/or at least one implicit relationship exists between any two of the plurality of concept definitions, synthesize a relationship between the two of the plurality of concept definitions.

13. The computer system of claim 12, wherein the at least one processor further executes the instructions to:

create a dimensional concept hierarchy, based at least in part on the synthesized relationships between concept definitions; and

classify content of the domain of information based on the dimensional concept hierarchy.

14. A computer storage product storing instructions that, when executed on a computer system, perform a method for generating a plurality of concept definitions through multiple tiers of abstraction using a domain of information, wherein the domain of information is a faceted domain that comprises a plurality of facets each having facet attributes, the method comprising:

extracting a plurality of concepts from the domain of information;

for each concept of the plurality of concepts, extracting at least one atomic concept associated with the concept from the domain of information, to extract a plurality of atomic concepts;

for each atomic concept of the plurality of atomic concepts, extracting at least one facet attribute associated with the atomic concept, to extract a plurality of facet attributes; and

generating a plurality of concept definitions, each of the plurality of concept definitions corresponding to a concept of the plurality of concepts, in accordance with a multi-tiered knowledge representation model, the multi-tiered knowledge representation model comprising at least three tiers, wherein a first tier of the multi-tiered knowledge representation model corresponds to the concepts, a second tier of the multi-tiered knowledge representation model corresponds to the atomic concepts, and a third tier of the multi-tiered knowledge representation model corresponds to the facet attributes, the generating comprising including at least one of the plurality of facet attributes in each of the concept definitions, in accordance with the associations between the extracted concepts, atomic concepts and facet attributes.

15. The computer storage product of claim 14, wherein the atomic concepts correspond to keywords, and the facet attributes correspond to morphemes.

16. The computer storage product of claim 14, wherein the method further comprises inferring facet attribute relationships from the plurality of concept definitions and concept relationships obtained from the domain of information.

17. The computer storage product of claim 14, wherein the method further comprises revising at least some of the plurality of concept definitions based at least in part on relationships between facet attributes of the at least some of the plurality of concept definitions through the multi-tiered knowledge representation model.

18. The computer storage product of claim 14, wherein the method further comprises:

determining whether any explicit relationships exist between the plurality of concept definitions, wherein an explicit relationship is determined to exist between any two of the plurality of concept definitions if the two of the plurality of concept definitions share at least one common facet attribute or each has a facet attribute of the same lineage in at least one facet attribute hierarchy of the plurality of facet attributes;

determining whether any implicit relationships exist between the plurality of concept definitions, wherein an implicit relationship between any two of the plurality of concept definitions is determined to exist based on a statistical identification of a relationship between a facet attribute in a first of the two of the plurality of concept definitions and a facet attribute in a second of the two of the plurality of concept definitions; and

when it is determined that at least one explicit relationship and/or at least one implicit relationship exists between any two of the plurality of concept definitions, synthesizing a relationship between the two of the plurality of concept definitions.

19. The computer storage product of claim 18, wherein the method further comprises creating a dimensional concept hierarchy, based at least in part on the synthesized relationships between concept definitions.

20. The computer storage product of claim 19, wherein the method further comprises classifying content of the domain of information based on the dimensional concept hierarchy.