We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Improving Scalable K-Means++.
- Authors
Hämäläinen, Joonas; Kärkkäinen, Tommi; Rossi, Tuomo
- Abstract
Two new initialization methods for K-means clustering are proposed. Both proposals are based on applying a divide-and-conquer approach for the K-means‖ type of an initialization strategy. The second proposal also uses multiple lower-dimensional subspaces produced by the random projection method for the initialization. The proposed methods are scalable and can be run in parallel, which make them suitable for initializing large-scale problems. In the experiments, comparison of the proposed methods to the K-means++ and K-means‖ methods is conducted using an extensive set of reference and synthetic large-scale datasets. Concerning the latter, a novel high-dimensional clustering data generation algorithm is given. The experiments show that the proposed methods compare favorably to the state-of-the-art by improving clustering accuracy and the speed of convergence. We also observe that the currently most popular K-means++ initialization behaves like the random one in the very high-dimensional cases.
- Subjects
RANDOM projection method; K-means clustering
- Publication
Algorithms, 2021, Vol 14, Issue 1, p6
- ISSN
1999-4893
- Publication type
Article
- DOI
10.3390/a14010006