Learning the attribute selection measures for decision tree

Xiaolin Chen; Zhihua Cai; Jia Wu

doi:10.1117/12.2021251

Learning the attribute selection measures for decision tree

Xiaolin Chen, Zhihua Cai, Jia Wu

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

1 Citation (Scopus)

Abstract

Decision tree has most widely used for classification. However the main influence of decision tree classification performance is attribute selection problem. The paper considers a number of different attribute selection measures and experimentally examines their behavior in classification. The results show that the choice of measure doesn’t affect the classification accuracy, but the size of the tree is influenced significantly. The main effect of the new attribute selection measures which base on normal gain and distance is that they generate smaller trees than traditional attribute selection measures.

Original language	English
Title of host publication	Fifth International Conference on Machine Vision (ICMV 2012)
Subtitle of host publication	algorithms, pattern recognition, and basic technologies
Editors	Yulin Wang, Liansheng Tan, Jianhong Zhou
Publisher	SPIE
Pages	87842S
Number of pages	8
Volume	8784
ISBN (Electronic)	9780819495884
DOIs	https://doi.org/10.1117/12.2021251
Publication status	Published - 2013
Externally published	Yes
Event	5th International Conference on Machine Vision (ICMV 2012) - Wuhan, China Duration: 20 Oct 2012 → 21 Oct 2012

Publication series

Name	Proceedings of SPIE
Publisher	SPIE
Volume	8784
ISSN (Electronic)	0277-786X

Conference

Conference	5th International Conference on Machine Vision (ICMV 2012)
Country	China
City	Wuhan
Period	20/10/12 → 21/10/12

Access to Document

10.1117/12.2021251

Fingerprint Dive into the research topics of 'Learning the attribute selection measures for decision tree'. Together they form a unique fingerprint.

Cite this

@inproceedings{43d3ad1d64c2496e89422e8d07ca684e,

title = "Learning the attribute selection measures for decision tree",

abstract = "Decision tree has most widely used for classification. However the main influence of decision tree classification performance is attribute selection problem. The paper considers a number of different attribute selection measures and experimentally examines their behavior in classification. The results show that the choice of measure doesn’t affect the classification accuracy, but the size of the tree is influenced significantly. The main effect of the new attribute selection measures which base on normal gain and distance is that they generate smaller trees than traditional attribute selection measures.",

author = "Xiaolin Chen and Zhihua Cai and Jia Wu",

year = "2013",

doi = "10.1117/12.2021251",

language = "English",

volume = "8784",

series = "Proceedings of SPIE",

publisher = "SPIE",

pages = "87842S",

editor = "Yulin Wang and Liansheng Tan and Jianhong Zhou",

booktitle = "Fifth International Conference on Machine Vision (ICMV 2012)",

address = "United States",

}

Chen, X, Cai, Z & Wu, J 2013, Learning the attribute selection measures for decision tree. in Y Wang, L Tan & J Zhou (eds), Fifth International Conference on Machine Vision (ICMV 2012): algorithms, pattern recognition, and basic technologies. vol. 8784, Proceedings of SPIE, vol. 8784, SPIE, pp. 87842S, 5th International Conference on Machine Vision (ICMV 2012), Wuhan, China, 20/10/12. https://doi.org/10.1117/12.2021251

Learning the attribute selection measures for decision tree. / Chen, Xiaolin; Cai, Zhihua; Wu, Jia.

Fifth International Conference on Machine Vision (ICMV 2012): algorithms, pattern recognition, and basic technologies. ed. / Yulin Wang; Liansheng Tan; Jianhong Zhou. Vol. 8784 SPIE, 2013. p. 87842S (Proceedings of SPIE; Vol. 8784).

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

TY - GEN

T1 - Learning the attribute selection measures for decision tree

AU - Chen, Xiaolin

AU - Cai, Zhihua

AU - Wu, Jia

PY - 2013

Y1 - 2013

N2 - Decision tree has most widely used for classification. However the main influence of decision tree classification performance is attribute selection problem. The paper considers a number of different attribute selection measures and experimentally examines their behavior in classification. The results show that the choice of measure doesn’t affect the classification accuracy, but the size of the tree is influenced significantly. The main effect of the new attribute selection measures which base on normal gain and distance is that they generate smaller trees than traditional attribute selection measures.

AB - Decision tree has most widely used for classification. However the main influence of decision tree classification performance is attribute selection problem. The paper considers a number of different attribute selection measures and experimentally examines their behavior in classification. The results show that the choice of measure doesn’t affect the classification accuracy, but the size of the tree is influenced significantly. The main effect of the new attribute selection measures which base on normal gain and distance is that they generate smaller trees than traditional attribute selection measures.

U2 - 10.1117/12.2021251

DO - 10.1117/12.2021251

M3 - Conference proceeding contribution

VL - 8784

T3 - Proceedings of SPIE

SP - 87842S

BT - Fifth International Conference on Machine Vision (ICMV 2012)

A2 - Wang, Yulin

A2 - Tan, Liansheng

A2 - Zhou, Jianhong

PB - SPIE

ER -