Knowledge Simulation via Relationship Mapping and Network Science

Abstract

Knowledge representation began with logic and ontology from Aristotle and focuses on managing information via structured metadata about relationships. Tools evolved employing subset approximation, categorization, and computational analysis that enable human understanding and mathematical manipulation. System fidelity requires that relationship richness be kept proportional to information size and complexity. This paper introduces knowledge simulation (Ks) resulting in knowledge inference (Ki). Ks is based on network science principles rather than structured metadata. Ki suggests knowledge potential by relaxing requirements for human understanding but increasing capability for human interaction in directing computational analysis.