Clusters analysis is frequently defined as the problem of partitioning a collection of objects into groups of similar objects according to some numerical measure of similarity. A wide variety of methods for doing this have been available for quite some time. The field has been—and remains—very well documented in the open literature [see, e.g., the books by Anderberg (1973), Diday (1979), Jambu (1978), Jambu and Lebeau (1983), Sokal and Sneath (1963), and the surveys by Diday and Simon (1976), Duda and Hart (1973), Redner and Walker (1984), to cite a few]. However, permanent emergence of new technical requirements as well as recognition of the limitations of existing techniques continue fostering intensive research activity worldwide which is mirrored by a permanent flow of publications proposing new ideas and algorithms.