Clustering in a Multi-Agent Data Mining Environment

Abstract

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.

Best Clustering Configuration Metrics: Towards Multiagent Based Clustering

“…These findings are currently being incorporated into a MAS framework [6,7]. The techniques investigated sofar, and reported here, do not serve to find the best results in all cases and further investigation is therefore required, however the authors are greatly encouraged by the result reported in this paper.…”

“…The process was incorporated into a MAC framework founded on earlier work by the authors and reported in [6,7]. An issue with K-means is that the initial points (records/objects) used to define the initial centroids of the clusters are randomly selected.…”

“…The need to be able to automatically identify best parameters with respect to a notion of a "best" cluster configuration is of particular relevance in the context of Multi Agent Clustering (MAC) as suggested in [6,7]. The view of clustering presented in [6] is that of a "anarchic" collection of agents: some equipped with algorithms to conduct clustering operations or to validate the output from clustering algorithms, some holding data, and others performing house keeping and management tasks.…”

“…The view of clustering presented in [6] is that of a "anarchic" collection of agents: some equipped with algorithms to conduct clustering operations or to validate the output from clustering algorithms, some holding data, and others performing house keeping and management tasks. The aim of this Multi-Agent System (MAS), although not fully realised, is to produce MAC solutions to clustering problems that require minimal input from the user and output a most appropriate cluster configuration.…”

A Multiple Pheromone Ant Clustering Algorithm

“…In many cases the algorithms do not have enough detail that allows them to be reverse engineered, and contact with their authors did not resolve the issue. References used in tables (4.4, 4.5) are as follows, (1) = [Handl et al, 2003a], (2) = [Tan et al, 2011], (3) = [Boryczka, 2010], (4) = , (5) = [Martens et al, 2007], (6) = [Xiong et al, 2012], (7) = [Chaimontree et al, 2010], (8) = [Breaban and Luchian, 2011], (9) = [Monmarché et al, 1999a], (10) = [Niknam and Amiri, 2010], (11) = [Santos and Bazzan, 2009], (12) = [Chandrasekar and Srinivasan, 2007], (13) = [Cano et al, 2013], (14) = [Wan et al, 2012], (15) = [Tan et al, 2006], (16) = [Bougenière et al, 2009], (17) = [Wang et al, 2007], (18) = [Jebara, 2002], (19) = [Rami and Panchal, 2012], (20) = [Azzag et al, 2007], (21) = [Labroche et al, 2002a], (22) = [Guo et al, 2003], (23) = [Yang and Zhang, 2007], (24) = [Ingaramo et al, 2005], (25) = [Shukran et al, 2011].…”