GENERAL FORMULATION AND EVALUATION OF AGGLOMERATIVE CLUSTERING METHODS WITH METRIC AND NONMETRIC DISTANCES

Agglomerative clustering methods with stopping criteria are generalize d. Clustering-related concepts are rigorously formulated with special consideration on metricity of object space. A new definition of combin atoriality is given, and a stronger proposition of monotonicity is pro ven. Specializations of the general method are applied to non-attribut ive non-metric and attributive pseudometric representations of biosequ ences. The furthest neighbor method is shown suitable for non-metric u se. In metric object space, four inter-clusteral distance functions, i ncluding a new truly context sensitive method, are compared using a me thod-independent goodness criterion. For biosequence clustering, the n ew method overcomes the UPGMA, UPGMC, and furthest neighbor methods.