SLINK: An optimally efficient algorithm for the single-link cluster method

Open Access
Publisher Website
Google Scholar
Abstract
The SLINK algorithm carries out single-link (nearest-neighbour) cluster analysis on an arbitrary dissimilarity coefficient and provides a representation of the resultant dendrogram which can readily be converted into the usual tree-diagram. The algorithm achieves the theoretical order-of-magnitude bounds for both compactness of storage and speed of operation, and makes the application of the single-link method feasible for a number of OTU's well into the range 103 to 104. The algorithm is easily programmable in a variety of languages including FORTRAN.