文档介绍：Automated Ontology Evolution in a Multi-Agent System
Mohsen Afsharchi, Behrouz H. Far
Department of Electrical & Computer Engineering
University of Calgary
Calgary, Canada
{mafsharc, far}***@
Abstract is to distribute the data repositories and enable
communication and message passing at the data level
This research addresses the formation of new by encapsulating the heterogeneity of legacy systems
concepts and their corresponding ontology in a multi- and applications within standard and interoperable
agent system where individual autonomous agents try wrappers. Interface description language and services,
which allow data objects to be defined, located and
to learn new concepts by consulting several other
agents. In this research individual agents create and invoked across locations and applications, are provided
learn their distinct conceptualization and rather than a for the puting environments [15].
commitment to mon ontology they use their own The most popular of such distributed paradigms are
OMG’mon Object Request Broker Architecture
ontologies. In this paper multi-agent supervised
learning of concepts among individual agents with (CORBA), Microsoft’s ponent Object
diverse conceptualization and different ontologies is Model (DCOM), and Sun’s Remote Method
introduced and demonstrated through an