Goal-driven semi-automated generation of semantic models

Abstract

The approach taken with OGEP is to parse relevant domain data in the form of unstructured content (or corpus) and use that knowledge to generate and/or evolve an existing ontology. OGEP creates a constant conversation between the corpus parser and a reasoning mechanism (corpus reasoner) that continually formulates potential ontology modifications in the form of hypotheses. These hypotheses are weighted towards contextual relevancy and further reasoned over to provide a confidence measure for use in deciding new assertions to the ontology. The new assertions generated from the corpus reasoner can either be automatically asserted based on confidence measure, or can be asserted by OGEP interacting with a user for final approval. This paper describes the OGEP technology in the context of the architectural components and identifies a potential technology transition path to Scott AFB's Tanker Airlift Control Center (TACC), which serves as the Air Operations Center (AOC) for the Air Mobility Command (AMC).