A primer on cluster analysis / Visual clustering methods

Jim Bezdek – Department of Electrical & Computer Engineering, University of Missouri, USA 

Part 1 gives definitions and notation associated with three types of clustering models: (i) prototype only = V models; (ii) partition only = U models; and (iii) joint (U, V) models. Each type is illustrated by one of its leading examples: (i) self-organizing maps; (ii) single linkage; and (iii) c-means. A fourth example given is the probabilistic mixture model, which is a (U, V) model with extra parameters. Part 1 is 100% tutorial, and is accessible to anyone with a little experience in computational mathematics. Fuzzy content ~ 25%; length = 90 minutes.Part 2 gives the definitions and notation associated with the three canonical problems of clustering: (i) pre-clustering tendency assessment; (ii) clustering, and (iii) post-clustering validation. A short history of visual clustering (which began in 1939) is followed by discussion of (8) algorithms developed by the author and various colleagues that address various facets of visual clustering. Part 2 is about 10% tutorial, and 90% specialized research. The objective is to present some state of the art research problems and solutions in the growing field of data visualization. Fuzzy content = none; length = 90 minutes.

Biography

Jim Bezdek received the PhD in Applied Mathematics from Cornell University in 1973. Jim is past president of NAFIPS (North American Fuzzy Information Processing Society), IFSA (International Fuzzy Systems Association) and the IEEE CIS (Computational Intelligence Society): founding editor the Int'l. Jo. Approximate Reasoning and the IEEE Transactions on Fuzzy Systems: fellow of the IEEE and IFSA; and a recipient of the IEEE 3rd Millennium, IEEE CIS Fuzzy Systems Pioneer, and IEEE CIS Rosenblatt medals. Jim's interests: woodworking, optimization, motorcycles, pattern recognition, cigars, clustering in very large data, fishing, poker, co-clustering, blues music, and visual clustering in relational data. Jim retired in 2007, and will be coming to a university near you soon.