Cross Split Decision Trees for pattern classification

Zahra Mirzamomen; Mohammad Navid Fekri; Mohammadreza Kangavari

doi:10.1109/ICCKE.2015.7365834

Cross Split Decision Trees for pattern classification

Zahra Mirzamomen, Mohammad Navid Fekri, Mohammadreza Kangavari

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

3 Citations (Scopus)

Abstract

One of the most important problems of decision trees is instability. It means that small changes in the dataset can result in different trees and different predictions. In this paper we introduce Cross Split Decision Tree (CSDT) which is a new decision tree learning algorithm with improved stability. This new algorithm uses multiple attributes as the split test in the internal nodes, in spite of the classical decision tree learning algorithms which use a single attribute. We have employed a heuristic based on the hoeffding bound to select the best attributes in the internal nodes. The experimental results show that in comparison with the well-known C4.5 decision tree learning algorithm, the proposed algorithm creates shallower decision trees with comparable accuracy.

Original language	English
Title of host publication	Proceedings of The 5th International Conference on Computer and Knowledge Engineering (ICCKE 2015)
Editors	Abbas Rasoolzadegan
Place of Publication	Piscataway NJ USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	240-245
Number of pages	6
ISBN (Electronic)	9781467392808
DOIs	https://doi.org/10.1109/ICCKE.2015.7365834
Publication status	Published - 2015
Externally published	Yes
Event	International Conference on Computer and Knowledge Engineering 2015 - Mashhad, Iran Duration: 29 Oct 2015 → 30 Oct 2020 Conference number: 5th https://ieeexplore.ieee.org/xpl/conhome/7360261/proceeding (Proceedings) http://iccke2015.um.ac.ir (Website)

Conference

Conference	International Conference on Computer and Knowledge Engineering 2015
Abbreviated title	ICCKE 2015
Country	Iran
City	Mashhad
Period	29/10/15 → 30/10/20
Internet address	https://ieeexplore.ieee.org/xpl/conhome/7360261/proceeding (Proceedings) http://iccke2015.um.ac.ir (Website)

Keywords

Decision Tree
Hoeffding Bound
Stability

Access to Document

10.1109/ICCKE.2015.7365834

Link to publication in Scopus

Cite this

@inproceedings{58e6222b96a34e4bbe8cf4213cd2d419,

title = "Cross Split Decision Trees for pattern classification",

abstract = "One of the most important problems of decision trees is instability. It means that small changes in the dataset can result in different trees and different predictions. In this paper we introduce Cross Split Decision Tree (CSDT) which is a new decision tree learning algorithm with improved stability. This new algorithm uses multiple attributes as the split test in the internal nodes, in spite of the classical decision tree learning algorithms which use a single attribute. We have employed a heuristic based on the hoeffding bound to select the best attributes in the internal nodes. The experimental results show that in comparison with the well-known C4.5 decision tree learning algorithm, the proposed algorithm creates shallower decision trees with comparable accuracy.",

keywords = "Decision Tree, Hoeffding Bound, Stability",

author = "Zahra Mirzamomen and Fekri, {Mohammad Navid} and Mohammadreza Kangavari",

year = "2015",

doi = "10.1109/ICCKE.2015.7365834",

language = "English",

pages = "240--245",

editor = "Rasoolzadegan, {Abbas }",

booktitle = "Proceedings of The 5th International Conference on Computer and Knowledge Engineering (ICCKE 2015)",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "International Conference on Computer and Knowledge Engineering 2015, ICCKE 2015 ; Conference date: 29-10-2015 Through 30-10-2020",

url = "https://ieeexplore.ieee.org/xpl/conhome/7360261/proceeding, http://iccke2015.um.ac.ir",

}

Mirzamomen, Z, Fekri, MN & Kangavari, M 2015, Cross Split Decision Trees for pattern classification. in A Rasoolzadegan (ed.), Proceedings of The 5th International Conference on Computer and Knowledge Engineering (ICCKE 2015). IEEE, Institute of Electrical and Electronics Engineers, Piscataway NJ USA, pp. 240-245, International Conference on Computer and Knowledge Engineering 2015, Mashhad, Iran, 29/10/15. https://doi.org/10.1109/ICCKE.2015.7365834

Cross Split Decision Trees for pattern classification. / Mirzamomen, Zahra; Fekri, Mohammad Navid; Kangavari, Mohammadreza.

Proceedings of The 5th International Conference on Computer and Knowledge Engineering (ICCKE 2015). ed. / Abbas Rasoolzadegan. Piscataway NJ USA : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 240-245.

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

TY - GEN

T1 - Cross Split Decision Trees for pattern classification

AU - Mirzamomen, Zahra

AU - Fekri, Mohammad Navid

AU - Kangavari, Mohammadreza

N1 - Conference code: 5th

PY - 2015

Y1 - 2015

N2 - One of the most important problems of decision trees is instability. It means that small changes in the dataset can result in different trees and different predictions. In this paper we introduce Cross Split Decision Tree (CSDT) which is a new decision tree learning algorithm with improved stability. This new algorithm uses multiple attributes as the split test in the internal nodes, in spite of the classical decision tree learning algorithms which use a single attribute. We have employed a heuristic based on the hoeffding bound to select the best attributes in the internal nodes. The experimental results show that in comparison with the well-known C4.5 decision tree learning algorithm, the proposed algorithm creates shallower decision trees with comparable accuracy.

AB - One of the most important problems of decision trees is instability. It means that small changes in the dataset can result in different trees and different predictions. In this paper we introduce Cross Split Decision Tree (CSDT) which is a new decision tree learning algorithm with improved stability. This new algorithm uses multiple attributes as the split test in the internal nodes, in spite of the classical decision tree learning algorithms which use a single attribute. We have employed a heuristic based on the hoeffding bound to select the best attributes in the internal nodes. The experimental results show that in comparison with the well-known C4.5 decision tree learning algorithm, the proposed algorithm creates shallower decision trees with comparable accuracy.

KW - Decision Tree

KW - Hoeffding Bound

KW - Stability

UR - http://www.scopus.com/inward/record.url?scp=84964466946&partnerID=8YFLogxK

U2 - 10.1109/ICCKE.2015.7365834

DO - 10.1109/ICCKE.2015.7365834

M3 - Conference Paper

AN - SCOPUS:84964466946

SP - 240

EP - 245

BT - Proceedings of The 5th International Conference on Computer and Knowledge Engineering (ICCKE 2015)

A2 - Rasoolzadegan, Abbas

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Piscataway NJ USA

T2 - International Conference on Computer and Knowledge Engineering 2015

Y2 - 29 October 2015 through 30 October 2020

ER -