Pruning multivariate decision trees by hyperplane merging

Miroslav Kubat, Doris Flotzinger

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Several techniques for induction of multivariate decision trees have been published in the last couple of years. Internal nodes of such trees typically contain binary tests questioning to what side of a hyperplane the example lies. Most of these algorithms use cut-off pruning mechanisms similar to those of traditional decision trees. Nearly unexplored remains the large domain of substitutional pruning methods, where a new decision test (derived from previous decision tests) replaces a subtree. This paper presents an approach to multivariate-tree pruning based on merging the decision hyperplanes, and demonstrates its performance on artificial and benchmark data.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages190-199
Number of pages10
Volume912
ISBN (Print)3540592865, 9783540592860
StatePublished - 1995
Externally publishedYes
Event8th European Conference on Machine Learning, ECML 1995 - Heraclion, Greece
Duration: Apr 25 1995Apr 27 1995

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume912
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other8th European Conference on Machine Learning, ECML 1995
CountryGreece
CityHeraclion
Period4/25/954/27/95

Fingerprint

Decision trees
Pruning
Merging
Decision tree
Hyperplane
Proof by induction
Binary
Benchmark
Internal
Vertex of a graph
Demonstrate

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Kubat, M., & Flotzinger, D. (1995). Pruning multivariate decision trees by hyperplane merging. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 912, pp. 190-199). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 912). Springer Verlag.

Pruning multivariate decision trees by hyperplane merging. / Kubat, Miroslav; Flotzinger, Doris.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 912 Springer Verlag, 1995. p. 190-199 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 912).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kubat, M & Flotzinger, D 1995, Pruning multivariate decision trees by hyperplane merging. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 912, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 912, Springer Verlag, pp. 190-199, 8th European Conference on Machine Learning, ECML 1995, Heraclion, Greece, 4/25/95.
Kubat M, Flotzinger D. Pruning multivariate decision trees by hyperplane merging. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 912. Springer Verlag. 1995. p. 190-199. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Kubat, Miroslav ; Flotzinger, Doris. / Pruning multivariate decision trees by hyperplane merging. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 912 Springer Verlag, 1995. pp. 190-199 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{6aa200aacdb14287b5f5842e40f3cfc7,
title = "Pruning multivariate decision trees by hyperplane merging",
abstract = "Several techniques for induction of multivariate decision trees have been published in the last couple of years. Internal nodes of such trees typically contain binary tests questioning to what side of a hyperplane the example lies. Most of these algorithms use cut-off pruning mechanisms similar to those of traditional decision trees. Nearly unexplored remains the large domain of substitutional pruning methods, where a new decision test (derived from previous decision tests) replaces a subtree. This paper presents an approach to multivariate-tree pruning based on merging the decision hyperplanes, and demonstrates its performance on artificial and benchmark data.",
author = "Miroslav Kubat and Doris Flotzinger",
year = "1995",
language = "English (US)",
isbn = "3540592865",
volume = "912",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "190--199",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Pruning multivariate decision trees by hyperplane merging

AU - Kubat, Miroslav

AU - Flotzinger, Doris

PY - 1995

Y1 - 1995

N2 - Several techniques for induction of multivariate decision trees have been published in the last couple of years. Internal nodes of such trees typically contain binary tests questioning to what side of a hyperplane the example lies. Most of these algorithms use cut-off pruning mechanisms similar to those of traditional decision trees. Nearly unexplored remains the large domain of substitutional pruning methods, where a new decision test (derived from previous decision tests) replaces a subtree. This paper presents an approach to multivariate-tree pruning based on merging the decision hyperplanes, and demonstrates its performance on artificial and benchmark data.

AB - Several techniques for induction of multivariate decision trees have been published in the last couple of years. Internal nodes of such trees typically contain binary tests questioning to what side of a hyperplane the example lies. Most of these algorithms use cut-off pruning mechanisms similar to those of traditional decision trees. Nearly unexplored remains the large domain of substitutional pruning methods, where a new decision test (derived from previous decision tests) replaces a subtree. This paper presents an approach to multivariate-tree pruning based on merging the decision hyperplanes, and demonstrates its performance on artificial and benchmark data.

UR - http://www.scopus.com/inward/record.url?scp=84948988824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948988824&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540592865

SN - 9783540592860

VL - 912

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 190

EP - 199

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -