Agent technology to model evolvable ontologies

Abstract

Ontology is becoming popular as a powerful way to structure the knowledge sources for complex systems. Modeling of evolvable ontologies turns out to be a research challenge for many decades. This paper presents how the agent technology can be used to update their ontologies while they are in execution. This solution has been designed to work on a primitive ontology at the beginning. Interaction among the agents updates the ontology as per the user intervention. Protégé and JADE have been used to implement the evolvable ontology and the agents respectively. The solution also uses mobile technology to synchronize the end user communication when the desired users are not connected on to the system at a required time. This amounts to minimize resource utilization by the agent, and also to enable notifications to the inactive team members in a timely manner. The system has been used to organize an exhibition, and shown the power of evolving ontology to communicate previously unseen aspects and also to incorporate emergent needs of a planning task in general. Our solution has been able to effectively update ontologies of end users before they come for a formal meeting.