We are interested in the greedy method to compute an hierarchical complete linkage clustering. There are two known methods for this problem, one having a running time of O(n3) with a space requirement of O(n) and one having a running time of O(n2 log n) with a space requirement of Θ(n2), where n is the number of points to be clustered. Both methods are not capable to handle large point sets. In this paper, we give an algorithm with a space requirement of O(n) which is able to cluster one million points in a day on current commodity hardware. © 2014 Springer International Publishing.
CITATION STYLE
Althaus, E., Hildebrandt, A., & Hildebrandt, A. K. (2014). A greedy algorithm for hierarchical complete linkage clustering. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8542 LNBI, pp. 25–34). Springer Verlag. https://doi.org/10.1007/978-3-319-07953-0_2
Mendeley helps you to discover research relevant for your work.