C4I Center at GMU - 2010 STIDS Conference Tutorial

Tuesday, October 26

Bill Andersen, Chris Schanck, Lynne Plettenberg Highfleet, Inc. presentation For over 10 years, Highfleet Inc. (formerly Ontology Works) has employed the discipline of ontology to solve pressing problems and provide advanced knowledge discovery and question-answering capabilities to the intelligence and defense industries. Our applications, implemented by our deductive database technologies, have been called mission-critical. In this tutorial, we demonstrate lessons learned and best practices for building applications that move semantics out of the backroom and into the main information flow of the enterprise.
09: 30 - 10:15	Introduction and Overview of Tools
	Abstract We will outline our experiences applying ontologies in a range of domains and provide a brief overview of the Highfleet Integrated Ontology Development Environment, Extensible Knowledge Server and API, which we will use to demonstrate a real-world implementation of the concepts covered in the tutorial.
10:15 - 12:00	Developing a User-facing Ontology Application, Soup to Nuts
	Abstract Ontological models are only as good as the application chains that implement them. We will walk through all the phases of developing an ontology-backed application, discussing best practices in ontology development, model testing, client application development, and system maintenance along the way. The result will be a user-facing GUI application backed by a logical reasoning engine and a substantial dataset.
12:00 - 02:00	Lunch Break
02:00 - 03:30	Semantically Integrating Legacy Databases
	Abstract Semantic technologies provide significant leverage for efficient and accurate integration of existing data holdings. We will demonstrate best practices for federating legacy data sources from a wide variety of formats based on their semantics and methods. The result will be an ontology application that provides real-time dynamic reasoning across several traditional databases.
03:30 - 04:00	Break
04:00 - 05:00	Reification, Security and Provenance
	Abstract Reification -- making language statements into objects of reference -- is a useful technique for implementing metadata in semantic applications. The utility of reification is limited, however, if metadata cannot be combined with inference at the object level. In this session we will talk about applications of metadata, to include access control -- specifically multi-level security -- and provenance and Highfleet's plan to support these capabilities in future versions of XKS.