Systems and methods for semantic concept definition and semantic concept relationship synthesis utilizing existing domain definitions

• US 8,943,016 B2
• Filed: 06/17/2013
• Issued: 01/27/2015
• Est. Priority Date: 08/29/2008
• Status: Active Grant

• Current Assignee: Primal Fusion Inc.
• Original Assignee: Primal Fusion Inc.
• Inventors: Sweeney, Peter, Black, Alexander David
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): OLUDE AFOLABI, OLATOYOSI
• Application Number:
  US13/919,934
• Publication Number:
  US 20130282647A1
• Time in Patent Office: 589 Days
• Field of Search: 706/55, 707/100
• US Class Current: 706/55
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition

Abstract

Computer-implemented systems and methods for synthesis of concept definitions and concept relationships from a domain of data, utilizing different semantic processing protocols such as formal concept analysis and faceted classification synthesis from existing domain concepts that have a confidence gradient built into them. A cognitive of an input agent provides an input of an active concept which is matched against existing domain concepts. The resultant pool of relevant domain concepts is then used to derive virtual concept definitions using a semantic processing protocol. The derivation is then overlaid with a concept of relative proximity of an attribute from another within an attribute set. An additional layer of coherence is given by the relative proximity measure. The end result is a pool of related virtual concept definitions in a tree structure.

Claims

1. A computer-implemented method for generating at least one concept tree comprising a plurality of concepts and concept relationships, said method comprising:

obtaining, either directly or indirectly via a network, on one or more processors executing software code, input information from a cognitive agent providing a basis for an active concept;

applying, based on the received input information, a semantic processing protocol to a domain of information comprising a data source from which at least one of the plurality of concepts derivable;

generating, from the application of the semantic protocol, from an extraction of concepts from documents, and/or from some combination thereof, the at least one concept tree that includes the plurality of concepts and concept relationships;

wherein at least one of the plurality of concepts has been analyzed for a level of coherence with the active concept;

wherein the cognitive agent includes at least one of a human user, a computer program receiving an input directly or indirectly from a human user, and/or some combination thereof; and

wherein the semantic protocol comprises at least one of concept analysis, data analysis, faceted classification synthesis, semantic reasoners, and/or some combination thereof.

2. The method of claim 1, wherein generating the at least one concept tree comprises extracting concepts from at least one of the input information, the domain of information, and/or some combination thereof.

3. The method of claim 1, wherein generating the at least one concept tree comprises assigning a priority level to at least two of the plurality of concepts.

4. The method of claim 1, wherein generating the at least one concept tree comprises representing the at least one concept tree in at least one of a directed or an undirected graph.

5. The method of claim 4, wherein generating the at least one concept tree comprises embodying the at least one concept tree in a knowledge representation comprising a semantic network.

6. The method of claim 1, wherein generating the at least one concept tree comprises executing the semantic processing protocol on a system that is remote from a system on which inputs are received from the cognitive agent.

7. The method of claim 1, wherein the data source included in the domain of information is a natural language data source.

8. The method of claim 1, wherein the data source included in the domain of information is a machine-readable data source generated at least in part by a computer application.

9. A computer-implemented method for generating at least one concept hierarchy comprising a plurality of concepts and concept relationships, said method comprising:

receiving, either directly or indirectly via a network, on one or more processors executing software code, input information from a cognitive agent providing a basis for an active concept;

applying, based on the received input information, a semantic processing protocol to a domain of information comprising a data source from which at least one of the plurality of concepts is derivable;

generating, from the application of the semantic protocol, from an extraction of concepts from documents, and/or from some combination thereof, the at least one concept hierarchy that includes the plurality of concepts and concept relationships;

wherein at least one of the plurality of concepts is determined to have a concept relationship with the active concept comprising at least one of a broader, a narrower, and/or an attributiorial relationship;

wherein the cognitive agent includes at least one of a human user, a computer program receiving an input directly or indirectly from a human user, and/or some combination thereof; and

wherein the semantic protocol comprises at least one of concept analysis, data analysis, faceted classification synthesis, semantic reasoners, and/or some combination thereof.

10. The method of claim 9, wherein generating the at least one concept hierarchy comprises forming a concept tree.

11. The method of claim 9, wherein the concept relationships are hierarchical and based on the domain of information.

12. The method of claim 9, wherein generating the at least one concept hierarchy comprises assembling a hierarchy through a semantic protocol.

13. The method of claim 9, wherein generating the at least one concept hierarchy comprises extracting concepts from the input information, the domain of information, and/or some combination thereof.

14. The method of claim 9, wherein generating the at least one concept hierarchy comprises assigning a priority level to at least one of the plurality of concepts.

15. The method of claim 9, wherein generating the at least one concept hierarchy comprises embodying the at least one concept hierarchy in a knowledge representation comprising a semantic network.

16. The method of claim 9, wherein generating the at least one concept hierarchy comprises executing a semantic protocol on a system that is remote from a system on which inputs are received from the cognitive agent.

17. The method of claim 9, wherein the data source included in the domain of information is a natural language data source.

18. The method of claim 9, wherein the data source included in the domain of information is a machine-readable data source generated at least in part by a computer application.

19. A computer system for generating at least one concept tree comprising a plurality of concepts and concept relationships, the system comprising:

at least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one processor, perform;

obtaining, either directly or indirectly via a network, on one or more processors executing software code, input information from a cognitive agent providing a basis for an active concept;

applying, based on the received input information, a semantic processing protocol to a domain of information comprising a data source from which at least one of the plurality of concepts is derivable;

generating, from the application of the semantic protocol, from an extraction of concepts from documents, and/or from some combination thereof, the at least one concept tree that includes the plurality of concepts and concept relationships;

wherein at least one of the plurality of concepts has been analyzed for a level of coherence with the active concept;

wherein the cognitive agent includes at least one of a human user, a computer program receiving an input directly or indirectly from a human user, and/or some combination thereof; and

wherein the semantic protocol comprises at least one of concept analysis, data analysis, faceted classification synthesis, semantic reasoners, and/or some combination thereof.