Data Retrieval Assistant for Consistent simulation data exploration using Natural Language interfaces (DRACO)

KBSI proposes to establish an innovative solution approach to support the analysis and enhanced understanding of federated simulation output data in a manner that consistently increases data analysts’ efficiency and effectiveness. Specifically, KBSI will research, design, and demonstrate the Data Retrieval Assistant for Consistent simulation data exploration using Natural Language interfaces (DRACO)’ system. DRACO will (i) accept natural language inputs, (ii) use semantic methods to auto-generate database queries, (iii) integrate the query results, and (iv) provide advanced and dynamic human machine interface (HMI) mechanisms to enable intuitive end user understanding and insight. The Phase I project will research, design, and demonstrate the technical feasibility of the DRACO solution approach. Phase II will refine, enhance, validate, and mature DRACO leading to a focused MDA application, leading to rapid technology transition and commercialization. Important innovations include (i) semantic disambiguation methods for processing Natural Language questions and commands, (ii) ontology based approach for automated query generation, (iii) rule based methods for query output fusion, (iv) dynamic and interactive human machine interfaces for enhanced sense making and human understanding, and (v) automated learning methods to enable adaptation of the search and exploration to different users and to dynamically changing BMDS data and ontologies. Approved for Public Release | 20-MDA-10398 (2 Mar 20)