Obtaining an optimal MAS configuration for agent-enhanced mining using constraint optimization

Authors:
Chayapol Moemeng;Can Wang;Longbing Cao
Affiliations:
Quantum Computing and Intelligent Systems, Faulty of Engineering and Information Technology, University of Technology, Sydney, Broadway, NSW, Australia;Quantum Computing and Intelligent Systems, Faulty of Engineering and Information Technology, University of Technology, Sydney, Broadway, NSW, Australia;Quantum Computing and Intelligent Systems, Faulty of Engineering and Information Technology, University of Technology, Sydney, Broadway, NSW, Australia
Venue:
ADMI'11 Proceedings of the 7th international conference on Agents and Data Mining Interaction
Year:
2011

Citing 9
Cited 0

Agent-enhanced Workflow

BT Technology Journal
Agents as Web Services

IEEE Internet Computing
Efficient Distributed Workflow Management Based on Variable Server Assignments

CAiSE '00 Proceedings of the 12th International Conference on Advanced Information Systems Engineering
Casting Mobile Agents to Workflow Systems: On Performance and Scalability Issues

DEXA '01 Proceedings of the 12th International Conference on Database and Expert Systems Applications
Towards Adaptive Workflow Enactment Using Multiagent Systems

Information Technology and Management
An efficient algorithm for multiagent plan coordination

Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems
Agent-based integration of Web Services with Workflow Management Systems

Proceedings of the fourth international joint conference on Autonomous agents and multiagent systems
Agent Mining: The Synergy of Agents and Data Mining

IEEE Intelligent Systems
Integrating workflow into agent-based distributed data mining systems

ADMI'10 Proceedings of the 6th international conference on Agents and data mining interaction

Quantified Score

Hi-index	0.00

Visualization

Abstract

We investigate an interaction mechanism between agents and data mining, and focus on agent-enhanced mining. Existing data mining tools use workflow to capture user requirements. The workflow enactment can be improved with a suitable underlying execution layer, which is a Multi-Agent System (MAS). From this perspective, we propose a strategy to obtain an optimal MAS configuration from a given workflow when resource access restrictions and communication cost constraints are concerned, which is essentially a constraint optimization problem. In this paper, we show how workflow is modeled in the way that can be optimized, and how the optimized model is used to obtain an optimal MAS configuration. Finally, we demonstrate that our strategy can improve the load balancing and reduce the communication cost during the workflow enactment.