Ontology Facades in Multi-Agent Systems for Logistic Applications: A Cluster-Based Approach
  • Agency: Office of Naval Research.
  • Duration: April 2001- November 2001.

   Capabilities Being Demonstrated:
  • Formulating a high-level specification language for an ontology
  • Implementing specialization features for an ontology
  • Integration of the ontology into an existing ontology framework

The focus of this project is to create a software support environment for building and reusing ontologies so that the cost of ontology design in multi-agent systems is amortized over several different projects. Even though the goal of an ontological engineer is to try and formulate the ontology in a general manner, in reality, the design issues in an ontology gets influenced by the need to solve the problem at hand in an optimal fashion. This implicit bias poses a problem for reuse of the ontology for subsequent projects, in which agents have to often deal with a slight shift in focus on the problem solving aspects, while using a domain similar to the previous project. Due to the existence of the slight slant in the current ontology's perspective, the new problem may get formulated at best, clumsily, in the old framework, and sometimes may not be able to get formulated at all. The good news is that ontological engineers can often provide insight into the types of modifications needed to render the old framework reusable for the new problem. However, the cost of understanding the complexities in the current ontology, recasting the new problem in the old framework and then deciding what changes should take place in the old ontology to effect a natural problem formulation, is an expensive proposition. The frequency with which old ontologies need to be recast as well as the extent to which they need to be recast, warrants that a high level approach be taken towards semi-automating some of the ontology redesign tasks.

In this project Pragati is building a prototype façade creation toolkit that allows a user to build and modify agent ontologies based on a high-level specification of the domain. If a high-level specification language can capture the agent interest succinctly in terms of the changes to be made to an existing or core ontology, then a set of filters, called façades, can be built around such a specification. A façade will allow the user to obtain his /her view of the world for the new agent's problem formulation, without actually changing the core ontology.

« Back | Top
©2007 Pragati Synergetic Research, Inc. Research Focus | Products | Projects | Publications | Market Sectors | About Us