Scaling-up model-based clustering algorithm by working on clustering features
Document Type
Conference paper
Source Publication
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Publication Date
1-1-2002
Volume
2412
First Page
569
Last Page
575
Publisher
Springer Verlag
Abstract
In this paper, we propose EMACF (Expectation- Maximization Algorithm for Clustering Features) to generate clusters from data summaries rather than data items directly. Incorporating with an adaptive grid-based data summarization procedure, we establish a scalable clustering algorithm: gEMACF. The experimental results show that gEMACF can generate more accurate results than other scalable clustering algorithms. The experimental results also indicate that gEMACF can run two order of magnitude faster than the traditional expectation-maximization algorithm with little loss of accuracy.
DOI
10.1007/3-540-45675-9_86
Print ISSN
03029743
Publisher Statement
Copyright © Springer-Verlag Berlin Heidelberg 2002. Access to external full text or publisher's version may require subscription.
Additional Information
Paper presented at the 3rd International Conference on Intelligent Data Engineering and Automated Learning, Aug 12-14, 2002, Manchester, England.
ISBN of the source publication: 9783540440253
Full-text Version
Publisher’s Version
Language
English