Toolkit for Agent-based Knowledge Extraction (TAKE)

We propose to research, develop, and validate an innovative Toolkit for Agent-based Knowledge Extraction (TAKE). The proposed research uses (a) a hybrid approach that combines ontology engineering methods with emerging knowledge discovery techniques to extract, analyze, and map ontologies from distributed and disparate knowledge and data sources and (b) deploys the ontology mapping results for the analysis and redesign of complex, multi-agent systems. The Phase I effort (i) established a novel TAKE ontology reuse and application method; (ii) designed innovative ontology analysis and ontology extraction algorithms; (iii) designed the TAKE architecture; and (iv) demonstrated a prototype of the TAKE software on focused, high payoff Navy transition application scenarios. The Phase II project will (a) refine and enhance the novel ontology reuse and application methods, (b) enhance the TAKE architecture, and (c) demonstrate the practical benefits of the innovation by building and validating two focused Navy TAKE applications leading to accelerated technology transition and commercialization. The main results of the Phase II effort will be (i) TAKE ontology similarity analysis and knowledge extraction methods, (ii) TAKE architecture, and (iii) two focused Navy applications.