Optimization of Decision Trees with Hypotheses for Knowledge Representation

Mohammad Azad; Igor Chikalov; Shahid Hussain; Mikhail Moshkov

doi:10.3390/electronics10131580

Optimization of Decision Trees with Hypotheses for Knowledge Representation

Mohammad Azad, Igor Chikalov, Shahid Hussain, Mikhail Moshkov

Research output: Contribution to journal › Article › peer-review

6 Scopus citations

Abstract

In this paper, we consider decision trees that use two types of queries: queries based on one attribute each and queries based on hypotheses about values of all attributes. Such decision trees are similar to the ones studied in exact learning, where membership and equivalence queries are allowed. We present dynamic programming algorithms for minimization of the depth and number of nodes of above decision trees and discuss results of computer experiments on various data sets and randomly generated Boolean functions. Decision trees with hypotheses generally have less complexity, i.e., they are more understandable and more suitable as a means for knowledge representation.

Original language	English (US)
Pages (from-to)	1580
Journal	Electronics
Volume	10
Issue number	13
DOIs	https://doi.org/10.3390/electronics10131580
State	Published - Jun 30 2021

Bibliographical note

KAUST Repository Item: Exported on 2021-07-02
Acknowledgements: Research reported in this publication was supported by King Abdullah University of Science and Technology (KAUST) including the provision of computing resources. The authors are greatly indebted to the anonymous reviewers for useful comments and suggestions.

Access to Document

10.3390/electronics10131580

https://repository.kaust.edu.sa/bitstream/10754/669869/1/electronics-10-01580.pdf

Cite this

@article{7976144e5fd546f6a5826d3b9e6f9492,

title = "Optimization of Decision Trees with Hypotheses for Knowledge Representation",

abstract = "In this paper, we consider decision trees that use two types of queries: queries based on one attribute each and queries based on hypotheses about values of all attributes. Such decision trees are similar to the ones studied in exact learning, where membership and equivalence queries are allowed. We present dynamic programming algorithms for minimization of the depth and number of nodes of above decision trees and discuss results of computer experiments on various data sets and randomly generated Boolean functions. Decision trees with hypotheses generally have less complexity, i.e., they are more understandable and more suitable as a means for knowledge representation.",

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

note = "KAUST Repository Item: Exported on 2021-07-02 Acknowledgements: Research reported in this publication was supported by King Abdullah University of Science and Technology (KAUST) including the provision of computing resources. The authors are greatly indebted to the anonymous reviewers for useful comments and suggestions.",

year = "2021",

month = jun,

day = "30",

doi = "10.3390/electronics10131580",

language = "English (US)",

volume = "10",

pages = "1580",

journal = "Electronics",

issn = "2079-9292",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "13",

}

TY - JOUR

T1 - Optimization of Decision Trees with Hypotheses for Knowledge Representation

AU - Azad, Mohammad

AU - Chikalov, Igor

AU - Hussain, Shahid

AU - Moshkov, Mikhail

N1 - KAUST Repository Item: Exported on 2021-07-02 Acknowledgements: Research reported in this publication was supported by King Abdullah University of Science and Technology (KAUST) including the provision of computing resources. The authors are greatly indebted to the anonymous reviewers for useful comments and suggestions.

PY - 2021/6/30

Y1 - 2021/6/30

N2 - In this paper, we consider decision trees that use two types of queries: queries based on one attribute each and queries based on hypotheses about values of all attributes. Such decision trees are similar to the ones studied in exact learning, where membership and equivalence queries are allowed. We present dynamic programming algorithms for minimization of the depth and number of nodes of above decision trees and discuss results of computer experiments on various data sets and randomly generated Boolean functions. Decision trees with hypotheses generally have less complexity, i.e., they are more understandable and more suitable as a means for knowledge representation.

AB - In this paper, we consider decision trees that use two types of queries: queries based on one attribute each and queries based on hypotheses about values of all attributes. Such decision trees are similar to the ones studied in exact learning, where membership and equivalence queries are allowed. We present dynamic programming algorithms for minimization of the depth and number of nodes of above decision trees and discuss results of computer experiments on various data sets and randomly generated Boolean functions. Decision trees with hypotheses generally have less complexity, i.e., they are more understandable and more suitable as a means for knowledge representation.

UR - http://hdl.handle.net/10754/669869

UR - https://www.mdpi.com/2079-9292/10/13/1580

U2 - 10.3390/electronics10131580

DO - 10.3390/electronics10131580

M3 - Article

SN - 2079-9292

VL - 10

SP - 1580

JO - Electronics

JF - Electronics

IS - 13

ER -

Optimization of Decision Trees with Hypotheses for Knowledge Representation

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this