Bernard Zeigler, Tucson, USA
| What | Invited talk |
|---|---|
| When |
2008-03-17 from 11:00 to 12:30 |
| Where | R 11, University Library, Albert-Einstein-Str.6 |
| Contact Name | Bernard Zeigler |
| Contact Email | zeigler@ece.arizona.edu |
| Add event to calendar |
|
Ontology Framework for Modeling and Simulation in Complex System Development
Bernard Zeigler
Arizona
Center for Integrative Modeling and
Simulation
University
of Arizona
Tucson, Arizona, USA
Ontologies are models of the world relating to specific aspects or applications that are typically represented in frame-based languages and form the knowledge components for logical agents on the Semantic Web. A convergence is underway that re-enforces the commonality of the object-based origins of artificial intelligence, knowledge representation, and software engineering. The Unified Modeling Language (UML) is being extended to incorporate ontology representations so that software systems in general will have more explicit models of their domains of operation. However, such a framework lacks several fundamental features that are particularly relevant to Modeling and Simulation (M&S) methodology. In this talk, we point out these limitations and then go on to discuss recent extensions to the System Entity Structure, (SES) a knowledge representation formalism that focuses on elements and relationships that are fundamental to model composition, validation, reuse, and interoperability. We show how the SES specifies a family of hierarchical, modular simulation models and can be viewed as an ontology for M&S with the set of all simulation models as its domain of discourse. Since simulation models include both static and dynamic elements with application to numerous real-world domains, the SES represents an advanced ontology framework. In particular, we discuss applications to automated model-driven development and testing of command, control and information systems.
Background for the talk can be found in the book:

