Identifying Groups: A Comparison of Methodologies
Abdolreza Eshghi, Dominique Haughton, Pascal Legrand
Abstract
This paper describes and compares three clustering techniques: traditional clustering methods, Kohonen maps and latent class models. The paper also proposes some novel measures of the quality of a clustering. To the best of our knowledge, this is the first contribution in the literature to compare these three techniques in a context where the classes are not known in advance.
Topics & Concepts
Cluster analysisComputer scienceContext (archaeology)Data miningClass (philosophy)Self-organizing mapArtificial intelligenceMachine learningGeographyArchaeologyData Mining Algorithms and ApplicationsAdvanced Clustering Algorithms ResearchBayesian Methods and Mixture Models