Clustering is a popular data mining technique, with applications in many areas. Although there are many clustering algorithms, none of them is superior on all datasets. Typically these clustering algorithms while providing summary statistics on the generated set of clusters do not provide easily interpretable detailed descriptions of the set of clusters that are generated. Further for a given dataset, different algorithms may give different sets of clusters, and so it is never clear which algorithm and which parameter settings is the most appropriate. In this paper we propose the use of a decision tree (DT) based approach that involves the use of multiple performance measures for indirectly assessing cluster quality in order to determine the most appropriate set of clusters. © 2005 Springer Science+Business Media, Inc.
CITATION STYLE
Osei-Bryson, K. M. (2005). Assessing cluster quality using multiple measures - a decision tree based approach. Operations Research/ Computer Science Interfaces Series, 29, 371–384. https://doi.org/10.1007/0-387-23529-9_24
Mendeley helps you to discover research relevant for your work.