The Community for Technology Leaders
Fourth IEEE International Conference on Data Mining (ICDM'04) (2004)
Brighton, United Kingdom
Nov. 1, 2004 to Nov. 4, 2004
ISBN: 0-7695-2142-8
pp: 67-74
Chris Giannella , University of Maryland Baltimore County, Baltimore, MD
Kun Liu , University of Maryland Baltimore County, Baltimore, MD
Todd Olsen , University of Maryland Baltimore County, Baltimore, MD
Hillol Kargupta , University of Maryland Baltimore County, Baltimore, MD
ABSTRACT
We present an algorithm designed to efficiently construct a decision tree over heterogeneously distributed data without centralizing. We compare our algorithm against a standard centralized decision tree implementation in terms of accuracy as well as the communication complexity. Our experimental results show that by using only 20% of the communication cost necessary to centralize the data we can achieve trees with accuracy at least 80% of the trees produced by the centralized version.
INDEX TERMS
Decision Trees, Distributed Data Mining, Random Projection
CITATION

C. Giannella, H. Kargupta, T. Olsen and K. Liu, "Communication Efficient Construction of Decision Trees Over Heterogeneously Distributed Data," Fourth IEEE International Conference on Data Mining (ICDM'04)(ICDM), Brighton, United Kingdom, 2004, pp. 67-74.
doi:10.1109/ICDM.2004.10114
85 ms
(Ver 3.3 (11022016))