Construction of Optimal Decision Trees and Deriving Decision Rules from Them

Mohammad Azad; Igor Chikalov; Shahid Hussain; Mikhail Moshkov; Beata Zielosko

doi:10.1007/978-3-031-08585-7_4

Construction of Optimal Decision Trees and Deriving Decision Rules from Them

Mohammad Azad, Igor Chikalov, Shahid Hussain, Mikhail Moshkov, Beata Zielosko

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

In this chapter, we propose dynamic programming algorithms for the construction of decision trees with minimum depth and decision trees with minimum number of working (nonterminal) nodes. We make computer experiments on various data sets from the UCI Machine Learning Repository and randomly generated Boolean functions to compare the length and coverage of decision rules derived from different kinds of optimal decision trees. The obtained results show that the decision rules derived from optimal decision trees with hypotheses in many cases are better than the rules derived from optimal conventional decision trees.

Original language	English (US)
Title of host publication	Decision Trees with Hypotheses
Publisher	Springer International Publishing
Pages	41-53
Number of pages	13
ISBN (Print)	9783031085840
DOIs	https://doi.org/10.1007/978-3-031-08585-7_4
State	Published - Nov 19 2022

Bibliographical note

KAUST Repository Item: Exported on 2022-12-02

Access to Document

10.1007/978-3-031-08585-7_4

Cite this

@inbook{fa8a039412be4ac78f1ff46c5a7c5f4f,

title = "Construction of Optimal Decision Trees and Deriving Decision Rules from Them",

abstract = "In this chapter, we propose dynamic programming algorithms for the construction of decision trees with minimum depth and decision trees with minimum number of working (nonterminal) nodes. We make computer experiments on various data sets from the UCI Machine Learning Repository and randomly generated Boolean functions to compare the length and coverage of decision rules derived from different kinds of optimal decision trees. The obtained results show that the decision rules derived from optimal decision trees with hypotheses in many cases are better than the rules derived from optimal conventional decision trees.",

author = "Mohammad Azad and Igor Chikalov and Shahid Hussain and Mikhail Moshkov and Beata Zielosko",

note = "KAUST Repository Item: Exported on 2022-12-02",

year = "2022",

month = nov,

day = "19",

doi = "10.1007/978-3-031-08585-7_4",

language = "English (US)",

isbn = "9783031085840",

pages = "41--53",

booktitle = "Decision Trees with Hypotheses",

publisher = "Springer International Publishing",

}

TY - CHAP

T1 - Construction of Optimal Decision Trees and Deriving Decision Rules from Them

AU - Azad, Mohammad

AU - Chikalov, Igor

AU - Hussain, Shahid

AU - Moshkov, Mikhail

AU - Zielosko, Beata

N1 - KAUST Repository Item: Exported on 2022-12-02

PY - 2022/11/19

Y1 - 2022/11/19

N2 - In this chapter, we propose dynamic programming algorithms for the construction of decision trees with minimum depth and decision trees with minimum number of working (nonterminal) nodes. We make computer experiments on various data sets from the UCI Machine Learning Repository and randomly generated Boolean functions to compare the length and coverage of decision rules derived from different kinds of optimal decision trees. The obtained results show that the decision rules derived from optimal decision trees with hypotheses in many cases are better than the rules derived from optimal conventional decision trees.

AB - In this chapter, we propose dynamic programming algorithms for the construction of decision trees with minimum depth and decision trees with minimum number of working (nonterminal) nodes. We make computer experiments on various data sets from the UCI Machine Learning Repository and randomly generated Boolean functions to compare the length and coverage of decision rules derived from different kinds of optimal decision trees. The obtained results show that the decision rules derived from optimal decision trees with hypotheses in many cases are better than the rules derived from optimal conventional decision trees.

UR - http://hdl.handle.net/10754/686111

UR - https://link.springer.com/10.1007/978-3-031-08585-7_4

U2 - 10.1007/978-3-031-08585-7_4

DO - 10.1007/978-3-031-08585-7_4

M3 - Chapter

SN - 9783031085840

SP - 41

EP - 53

BT - Decision Trees with Hypotheses

PB - Springer International Publishing

ER -

Construction of Optimal Decision Trees and Deriving Decision Rules from Them

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this