A versatile clustering algorithm with objective function and objective measure.

Abstract A computer program for nonparametric cluster synthesis, using similarity rather than maximum likelihood as the basis for class membership, is presented. The algorithm utilizes recursive computations to develop a hierarchy or tree of nested clusters. The major components of the program are: (1) a (dis)similarity function. (2) a grouping or merger strategy, based on optimizing a dynamic objective function, and (3) a halting criterion, based on evaluating a dynamic objective measure. Program options permit variations of data normalization, measures of similarity, and clustering strategy. A variety of hard-copy summaries and displays are available to the user. An illustrative application to the classification of human mitotic chromosomes is included.