US 12,032,616 B2
Systems and methods for semantic concept definition and semantic concept relationship synthesis utilizing existing domain definitions
Peter Sweeney, Kitchener (CA); and Alexander David Black, Guelph (CA)
Assigned to Primal Fusion Inc., Kitchener (CA)
Filed by Primal Fusion Inc., Kitchener (CA)
Filed on Jan. 30, 2017, as Appl. No. 15/418,875.
Application 15/418,875 is a continuation of application No. 14/571,902, filed on Dec. 16, 2014, granted, now 9,595,004.
Application 14/571,902 is a continuation of application No. 13/919,934, filed on Jun. 17, 2013, granted, now 8,943,016, issued on Jan. 27, 2015.
Application 13/919,934 is a continuation of application No. 12/549,812, filed on Aug. 28, 2009, granted, now 8,495,001, issued on Jul. 23, 2013.
Claims priority of provisional application 61/092,973, filed on Aug. 29, 2008.
Prior Publication US 2017/0140044 A1, May 18, 2017
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/36 (2019.01); G06F 40/30 (2020.01); G06N 5/022 (2023.01)
CPC G06F 16/36 (2019.01) [G06N 5/022 (2013.01); G06F 40/30 (2020.01)] 15 Claims
OG exemplary drawing
 
14. A system for analyzing synthesized concepts using a semantic processing protocol, wherein the synthesized concepts comprise a plurality of new virtual concept definitions derived from a set of real concept definitions by semantic processing, the system comprising:
at least one processor, operable to execute executable instructions stored in at least one tangible memory, to:
receive an input active concept definition from a cognitive agent via a system interface, 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 a combination thereof;
extract a plurality of real concept definitions comprising attributes from at least one domain, wherein the domain is a corpus of data representing a knowledge domain;
match the active concept definition to the extracted real concept definitions;
synthesize a plurality of new virtual concept definitions from the real concept definitions by semantic processing, such that the virtual concept definitions form relationships between themselves, wherein the virtual concept definitions have a tree structure; and
analyze the plurality of real concept definitions for attribute set coherence within the plurality of new virtual concept definitions, and the real concept definitions within a context of the active concept definition, wherein analysis of the synthesized virtual concept definitions is determined based on a confidence gradient derived from the attribute set coherence.