Optimization Clustering Techniques
Barry J. Everitt, Sabine Landau, Morven Leese, Daniel R. Stahl · Wiley series in probability and statistics · 2011
This chapter considers a class of clustering techniques which produces a partition of the individuals into a specified number of groups, by either minimizing or maximizing some numerical criterion. The basic idea behind the methods to be described in the chapter is that associated with each partition of the n individuals into the required number of groups, g, is an index c(n, g), the value of which measures some aspect of the “quality” of this particular partition. The chapter introduces cluster criteria derived from a dissimilarity matrix followed by criteria derived directly from continuous variables, and then discusses algorithms that can be used to optimize these criteria. Finally, the chapter presents several examples of applications of cluster optimization methods. Controlled Vocabulary Terms cluster analysis