Geometric algorithms for agglomerative hierarchical clustering

Authors:
Danny Z. Chen;Bin Xu
Affiliations:
Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN;Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN
Venue:
COCOON'03 Proceedings of the 9th annual international conference on Computing and combinatorics
Year:
2003

Citing 3
Cited 0

Euclidean minimum spanning trees and bichromatic closest pairs

Discrete & Computational Geometry
A decomposition of multidimensional point sets with applications to k-nearest-neighbors and n-body potential fields

Journal of the ACM (JACM)
Dealing with higher dimensions: the well-separated pair decomposition and its applications

Dealing with higher dimensions: the well-separated pair decomposition and its applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Agglomerative Hierarchical Clustering (AHC) methods play an important role in the science of classification. In this paper, We present exact centroid and median AHC algorithms and an approximate single-link AHC algorithm for clustering data objects in the d-D space for any constant integer d ≥ 2; the time and space bounds of all our algorithms are O(n log n) and O(n), where n is the size of the input data set. Previously best algorithmic approaches for these three methods take at least quadratic time in the worst case. We implemented these AHC algorithms, and the experimental results show that our algorithms are quite efficient and practical.