Abstract. Multi-Agent Clustering (MAC) requires a mechanism for identifying the most appropriate cluster configuration. This paper reports on experiments conducted with respect to a number of validation metrics to identify the most effective metric with respect to this context. This paper also describes a process whereby such metrics can be used to determine the optimum parameters typically required by clustering algorithms, and a process for incorporating this into a MAC framework to generate best cluster configurations with minimum input from end users.

Abstract. A framework for multi-agent based clustering is described whereby individual agents represent individual clusters. A particular feature of the framework is that, after an initial cluster configuration has been generated, the agents are able to negotiate with a view to improving on this initial clustering. The framework can be used in the context of a number of clustering paradigms, two are investigated: K-means and KNN. The reported evaluation demonstrates that negotiation can serve to improve on an initial cluster configuration.

A framework to support Multi-Agent Based Clustering (MABC) is described. A unique feature of the framework is that it provides mechanisms to allow agents to negotiate so as to improve an initial cluster configuration. The framework encourages a two phase approach to clustering. During the first phase clustering agents bid for records in the input data and form an initial cluster configuration. In the second phase (the negotiation phase) agents pass individual records to each other so as to improve the initial configuration. The communication framework and its operation is fully described in terms of the performatives used and from an algorithmic perspective. The reported evaluation was conducted using benchmark data sets. The results demonstrate that the supported agent negotiation produces enhanced clustering results.

Abstract.A framework for Multi Agent Data Mining (MADM) is described. The framework comprises a collection of agents cooperating to address given data mining tasks. The fundamental concept underpinning the framework is that it should support generic data mining. The vision is that of a system that grows in an organic manner. The central issue to facilitating this growth is the communication medium required to support agent interaction. This issue is partly addressed by the nature of the proposed architecture and partly through an extendable ontology; both are described. The advantages offered by the framework are illustrated in this paper by considering a clustering application. The motivation for the latter is that no "best" clustering algorithm has been identified, and consequently an agent-based approach can be adopted to identify "best" clusters. The application serves to demonstrates the full potential of MADM.

Abstract. A Multi-Agent based approach to clustering using a generic Multi-Agent Data Mining (MADM) framework is described. The process use a collection of agents, running several different clustering algorithms, to determine a "best" cluster configuration. The issue of determining the most appropriate configuration is a challenging one, and is addressed in this paper by considering two metrics, total Within Group Average Distance (WGAD) to determine cluster cohesion, and total Between Group Distance (BGD) to determine separation. The proposed process is implemented using the MASminer MADM framework which is also introduced in this paper. Both the clustering technique and MASminer are evaluated. Comparison of the two "best fit" measures indicates that WGAD can be argued to be the most appropriate metric.