Decision tree learning system with switching evaluator

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

Abstract

In this paper, we introduce the notion of the local strategy of constructing decision trees that includes the information theoretic entropy algorithm in ID3 (or C4.5) and any other local algorithms. Simply put, given a sample, a local algorithm constructs a decision tree in the top-down manner using an evaluation function. We propose a new local algorithm that is very different from the entropy algorithm. We analyze behaviors of the two algorithms on a simple model. Based on these analyses, we propose a learning system of decision trees which can change an evaluation function while constructing decision trees, and verify the effect of the system by experiments with real databases. The system not only achieves a high accuracy, but also produces well-balanced decision trees, which have the advantage of fast classification.

Original languageEnglish
Title of host publicationAdvances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings
PublisherSpringer Verlag
Pages349-361
Number of pages13
Volume1081
ISBN (Print)3540612912, 9783540612919
DOIs
Publication statusPublished - 1996
Externally publishedYes
Event11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996 - Toronto, Canada
Duration: 1996 May 211996 May 24

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1081
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996
CountryCanada
CityToronto
Period96/5/2196/5/24

Fingerprint

Learning Systems
Decision trees
Decision tree
Learning systems
Local Algorithms
Evaluation Function
Function evaluation
Entropy
High Accuracy
Verify
Experiment
Experiments

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Koshiba, T. (1996). Decision tree learning system with switching evaluator. In Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings (Vol. 1081, pp. 349-361). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1081). Springer Verlag. https://doi.org/10.1007/3-540-61291-2_64

Decision tree learning system with switching evaluator. / Koshiba, Takeshi.

Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings. Vol. 1081 Springer Verlag, 1996. p. 349-361 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1081).

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

Koshiba, T 1996, Decision tree learning system with switching evaluator. in Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings. vol. 1081, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1081, Springer Verlag, pp. 349-361, 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Toronto, Canada, 96/5/21. https://doi.org/10.1007/3-540-61291-2_64
Koshiba T. Decision tree learning system with switching evaluator. In Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings. Vol. 1081. Springer Verlag. 1996. p. 349-361. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-61291-2_64
Koshiba, Takeshi. / Decision tree learning system with switching evaluator. Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings. Vol. 1081 Springer Verlag, 1996. pp. 349-361 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{533664aead004ebebb90496a92e81f5b,
title = "Decision tree learning system with switching evaluator",
abstract = "In this paper, we introduce the notion of the local strategy of constructing decision trees that includes the information theoretic entropy algorithm in ID3 (or C4.5) and any other local algorithms. Simply put, given a sample, a local algorithm constructs a decision tree in the top-down manner using an evaluation function. We propose a new local algorithm that is very different from the entropy algorithm. We analyze behaviors of the two algorithms on a simple model. Based on these analyses, we propose a learning system of decision trees which can change an evaluation function while constructing decision trees, and verify the effect of the system by experiments with real databases. The system not only achieves a high accuracy, but also produces well-balanced decision trees, which have the advantage of fast classification.",
author = "Takeshi Koshiba",
year = "1996",
doi = "10.1007/3-540-61291-2_64",
language = "English",
isbn = "3540612912",
volume = "1081",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "349--361",
booktitle = "Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Decision tree learning system with switching evaluator

AU - Koshiba, Takeshi

PY - 1996

Y1 - 1996

N2 - In this paper, we introduce the notion of the local strategy of constructing decision trees that includes the information theoretic entropy algorithm in ID3 (or C4.5) and any other local algorithms. Simply put, given a sample, a local algorithm constructs a decision tree in the top-down manner using an evaluation function. We propose a new local algorithm that is very different from the entropy algorithm. We analyze behaviors of the two algorithms on a simple model. Based on these analyses, we propose a learning system of decision trees which can change an evaluation function while constructing decision trees, and verify the effect of the system by experiments with real databases. The system not only achieves a high accuracy, but also produces well-balanced decision trees, which have the advantage of fast classification.

AB - In this paper, we introduce the notion of the local strategy of constructing decision trees that includes the information theoretic entropy algorithm in ID3 (or C4.5) and any other local algorithms. Simply put, given a sample, a local algorithm constructs a decision tree in the top-down manner using an evaluation function. We propose a new local algorithm that is very different from the entropy algorithm. We analyze behaviors of the two algorithms on a simple model. Based on these analyses, we propose a learning system of decision trees which can change an evaluation function while constructing decision trees, and verify the effect of the system by experiments with real databases. The system not only achieves a high accuracy, but also produces well-balanced decision trees, which have the advantage of fast classification.

UR - http://www.scopus.com/inward/record.url?scp=84957927645&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84957927645&partnerID=8YFLogxK

U2 - 10.1007/3-540-61291-2_64

DO - 10.1007/3-540-61291-2_64

M3 - Conference contribution

AN - SCOPUS:84957927645

SN - 3540612912

SN - 9783540612919

VL - 1081

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

SP - 349

EP - 361

BT - Advances in Artificial Intelligence - 11th Biennial Conference of the Canadian Society for Computational Studies of Intelligence, AI 1996, Proceedings

PB - Springer Verlag

ER -