A COMPUTATIONAL MODEL FOR DISTRIBUTED KNOWLEDGE SYSTEMS WITH LEARNING-MECHANISMS

Authors
Citation
H. Aiba et T. Terano, A COMPUTATIONAL MODEL FOR DISTRIBUTED KNOWLEDGE SYSTEMS WITH LEARNING-MECHANISMS, Expert systems with applications, 10(3-4), 1996, pp. 417-427
Citations number
38
Categorie Soggetti
Operatione Research & Management Science","System Science","Engineering, Eletrical & Electronic","Computer Science Artificial Intelligence
ISSN journal
09574174
Volume
10
Issue
3-4
Year of publication
1996
Pages
417 - 427
Database
ISI
SICI code
0957-4174(1996)10:3-4<417:ACMFDK>2.0.ZU;2-8
Abstract
This paper addresses the issues of machine learning in distributed kno wledge systems, which will consist of distributed software agents with problem solving, communication and learning functions. To develop suc h systems, we must analyze the roles of problem-solving and communicat ion capabilities among knowledge systems. To facilitate the analyses, we propose a computational model: LPC. The model consists of a set of agents with (a) a knowledge base for learned concepts, (b) a knowledge base for problem solving, (c) prolog-based inference mechanisms and ( d) a set of beliefs on the reliability of the other agents. Each agent can improve its own problem-solving capabilities by deductive learnin g from the given problems, by memory-based learning from communication s between the agents and by reinforcement learning from the reliabilit y of communications between the other agents. An experimental system o f the model has been implemented in prolog language on a Window-based personal computer Intensive experiments have been carried out to exami ne the feasibility of the machine learning mechanisms of agents for pr oblem-solving and communication capabilities. The experimental results have shown that the multiagent system improves the performance of the whole system in problem solving, when each agent has a higher learnin g ability or when an agent with a very high ability for problem solvin g joins the organization to cooperate with the other agents in problem solving. These results suggest that the proposed model is useful in a nalyzing the learning mechanisms applicable to distributed knowledge s ystems. Copyright (C) 1996 Elsevier Science Ltd