Séminaire Donnees et APprentissage Artificiel


An inference algorithm based on mobile agents paradigm for distributed knowledge bases

Tuesday, November 2, 2010
Speaker(s) : Claudia Leon, Prof. à l'UCV de Caracas

In the 21st century's society knowledge is valuable only when it is widely available and used. Centralized systems accumulating knowledge do not ensure high performance features, as well as the advantages of collaborative work. Therefore, distributing knowledge seems to be logical choice. In this work we propose an algorithm that uses mobile agents as am integrating mechanism in the inference processes over distributed knowledge bases, represented by production rules. We uses mobile agents combined with production rules as they are easy way to represent knowledge, and they are used to study the mobile agents in the context of knowledge-based systems. The algorithm of distributed inference was implemented in Java, using JClips and Tracy2 tool. The algorithm was exhaustively tested on a local area network, to verify its correctness.

More details here …
Sahar.Changuel (at) nulllip6.fr