Two-neighborhood method for computing similarity between two groups of objects

摘要: A method for computing the similarity between two groups of objects wherein any pair can be computed by a function, being use in software clustering, when it is known that only are "near neighbors" to given object considered similarity, is, some positive integer "Y", Y-nearest neighbors considered, smaller "X" an same group as at least X its Y nearest neighbors, none other matter, and input comprises set containing objects, specific groups, B, values Y, comprising: (a) each object, identifying (hereafter called X-neighbors); (b) hereinafter Y-neighbors; (c) which neither Y-neighbor other, redefining them zero; (d) whose X-neighbors all itself, (e). outputing whereby maximum B.