Categorizing concept types of a conceptual graph

• US 9,087,293 B2
• Filed: 12/23/2008
• Issued: 07/21/2015
• Est. Priority Date: 12/23/2008
• Status: Active Grant

• Current Assignee: Forcepoint Federal Holdings LLC (Forcepoint LLC)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Peoples, Bruce E., Johnson, Michael R., Smith, Jonathon P.
• Primary Examiner(s): WILSON, KIMBERLY LOVEL
• Application Number:
  US12/342,580
• Publication Number:
  US 20100161669A1
• Time in Patent Office: 2,401 Days
• Field of Search: 707/736, 707/791
• US Class Current: 1/1
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation

Abstract

According to one embodiment, categorizing concept types of a conceptual graph includes receiving the conceptual graph comprising one or more concept types, one or more relationship types, and one or more arcs. Each concept type is categorized according to the relationship types and the arcs. The categorization of the each concept type is recorded.

Claims

1. A method implemented in a computer system comprising one or more processors in operable communication with one or more non-transitory, tangible, computer-readable storage media, at least a first one of the non-transitory, tangible computer-readable storage media storing instructions executable by at least one processor, the method comprising:

receiving, at the processor, a conceptual graph comprising one or more concept types, one or more relationship types, and one or more arcs;

causing the processor to execute one or more instructions configured to;

categorize each concept type of the one or more concept types according to the one or more relationship types and the one or more arcs, wherein the categorizing each concept type further comprises;

determining that each concept type is directly connected to two or more relationship types by two or more arcs pointing in different directions; and

categorizing the each concept type as a context linking concept;

record in a second respective one of the non-transitory, tangible, storage media, the categorization of the each concept type of the one or more concept types;

identify, based on a respective database of terms stored in a third respective one of the non-transitory, tangible storage media, one or more related terms of at least one particular concept type of the one or more concept types according to the categorization; and

searching a plurality of documents in a fourth respective one of the non-transitory, tangible, storage media, for matches to the identified one or more related terms of the at least one particular concept type according to the categorization.

2. The method of claim 1, the categorizing each concept type further comprising:

determining that each concept type fits a context linking concept pattern.

3. The method of claim 1, the categorizing each concept type further comprising:

determining that each concept type fits a concept object pattern; and

categorizing each concept type as a concept object.

4. The method of claim 1, the categorizing each concept type further comprising:

determining that each concept type is directly connected to one or more relationship types by one or more arcs pointing in the same direction; and

categorizing the each concept type as a concept object.

5. The method of claim 1, the recording the categorization further comprising:

mapping, to each related term, a category designation designating the category of the related term.

6. The method of claim 1, the recording the categorization further comprising:

recording the categorization in an onomasticon.

7. The method of claim 1, the recording the categorization further comprising:

mapping, for each related term, a category designation designating the category of each related term to a term semantically related to each related term.

8. The method of claim 1, the conceptual graph comprising a query conceptual graph generated from a search query.

9. The method of claim 1, wherein:

a document conceptual graph represents a first document of the plurality of documents;

the document conceptual graph comprises one or more document concept types; and

searching the plurality of document concept types further comprises;

identifying one or more related terms of a document concept type;

comparing the one or more related terms of the document concept type to the identified one or more related terms of the at least one particular concept type according to the categorization; and

selecting the first document as a result of a search based on the comparison of the one or more related terms of the document concept type to the identified one or more related terms of the at least one particular concept type according to the categorization.

10. A non-transitory, tangible, computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by one or more processors, are operable to:

receive a conceptual graph comprising one or more conceptual types, one or more relationship types, and one or more arcs;

categorize each concept type of the one or more concept types according to the one or more relationship types and the one or more arcs, the computer-executable instructions further configured to categorize each concept type by;

determining that each concept type is directly connected to two or more relationship types by two or more arcs pointing in different directions; and

categorizing the each concept type as a context linking concept;

record, in a non-transitory, tangible computer-readable memory device, the categorization of each concept type of the one or more concept types;

identify one or more related terms of at least one particular concept type of the one or more concept types according to the categorization; and

search a non-transitory, tangible computer-readable knowledgebase for a plurality of documents for matches to the identified one or more related terms of the at least one particular concept type according to the categorization.

11. The non-transitory, tangible, computer-readable storage medium of claim 10, the computer-executable instructions configured to categorize each concept type by:

determining that the each concept type fits a context linking concept pattern.

12. The non-transitory, tangible, computer-readable storage medium of claim 10, the computer-executable instructions further configured to categorize each concept type by:

determining that each concept type fits a concept object pattern; and

categorizing each concept type as a concept object.

13. The non-transitory, tangible, computer-readable storage medium of claim 10, the computer-executable instructions further configured to categorize each concept type by:

determining that each concept type is directly connected to one or more relationship types by one or more arcs pointing in the same direction; and

categorizing each concept type as a concept object.

14. The non-transitory, tangible, computer-readable storage mediums of claim 10, the computer-executable instructions further configured to record the categorization by:

mapping, to each related term, a category designation designating the category of the related term.

15. The non-transitory, tangible, computer-readable storage medium of claim 10, the computer-executable instructions further configured to record the categorization by:

recording the categorization in an onomasticon.

16. The non-transitory, tangible, computer-readable storage mediums of claim 10, the computer-executable instructions further configured to record the categorization by:

mapping, for each related term, a category designation designating the category of each related term to a term semantically related to each related term.

17. The non-transitory, tangible, computer-readable storage medium of claim 10, the conceptual graph comprising a query conceptual graph generated from a search query.

18. The non-transitory, tangible, computer-readable storage medium of claim 10, wherein:

a document conceptual graph represents a first document of the plurality of documents;

the document conceptual graph comprises one or more document concept types; and

the logic is further configured to search the plurality of documents by;

identifying one or more related terms of a document concept type;

comparing the one or more related terms of the document concept type to the identified one or more related terms of the at least one particular concept type according to the categorization; and

selecting the first document as a result of a search based on the comparison of the one or more related terms of the document concept type to the identified one or more related terms of the at least one particular concept type according to the categorization.