Empirical Evaluation of Clustering Algorithms.pdfVIP

1 Empirical Evaluation of Clustering Algorithms* Andreas Rauber¹, Jan Paralic², Elias Pampalk¹ ¹Vienna University of Technology, Department of Software Technology {andi, elias}@ifs.tuwien.ac.at ²Technical University of Kosice, Department of Cybernetics and Artificial Intelligen ce paralic@tuke.sk Abstract: Unsupervised data classification can be considered one of the most important initial steps in the process of data mining. Numerous algorithms have be en developed and are being used in this context in a variety of application domains. Albeit, only little evidence is available as to which algorithms shoul d be used in which context, and which techniques offer p romising results when being combined for a given ta sk. In this paper we present an empirical evaluation of some prominent unsupervised data classification te ch- niques with respect to their usability and the inte rpretability of their result representation. 1 INTRODUCTION Detecting unknown patterns in large datasets is bec oming an increasingly important factor with respect to the wid