Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data

Peerapon Vateekul; Miroslav Kubat

doi:10.1109/ICDMW.2009.94

Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data

Peerapon Vateekul, Miroslav Kubat

Electrical and Computer Engineering

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

13 Scopus citations

Abstract

The paper focuses on automated categorization of text documents, each labeled with one or more classes and described by tens of thousands of features. The computational costs of induction in such domains are so high as almost to disqualify the use of decision trees; the reduction of these costs is thus an important research issue. Our own solution, FDT ("fast decision-tree induction"), uses a two-pronged strategy: (1) feature-set pre-selection, and (2) induction of several trees, each from a different data subset, with the combination of the results from multiple trees with a data-fusion technique tailored to domains with imbalanced classes.

Original language	English (US)
Title of host publication	ICDM Workshops 2009 - IEEE International Conference on Data Mining
Pages	320-325
Number of pages	6
DOIs	https://doi.org/10.1109/ICDMW.2009.94
State	Published - Dec 1 2009
Event	2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009 - Miami, FL, United States Duration: Dec 6 2009 → Dec 6 2009

Publication series

Name	ICDM Workshops 2009 - IEEE International Conference on Data Mining

Other

Other	2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009
Country	United States
City	Miami, FL
Period	12/6/09 → 12/6/09

Keywords

Decision tree
Imbalanced classes
Large-scale data
Multi-label examples
Text categorization

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Vision and Pattern Recognition
Software

Access to Document

10.1109/ICDMW.2009.94

Fingerprint Dive into the research topics of 'Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data'. Together they form a unique fingerprint.

Cite this

Vateekul, P., & Kubat, M. (2009). Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data. In ICDM Workshops 2009 - IEEE International Conference on Data Mining (pp. 320-325). [5360425] (ICDM Workshops 2009 - IEEE International Conference on Data Mining). https://doi.org/10.1109/ICDMW.2009.94

Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data. / Vateekul, Peerapon; Kubat, Miroslav.

ICDM Workshops 2009 - IEEE International Conference on Data Mining. 2009. p. 320-325 5360425 (ICDM Workshops 2009 - IEEE International Conference on Data Mining).

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

Vateekul, P & Kubat, M 2009, Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data. in ICDM Workshops 2009 - IEEE International Conference on Data Mining., 5360425, ICDM Workshops 2009 - IEEE International Conference on Data Mining, pp. 320-325, 2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009, Miami, FL, United States, 12/6/09. https://doi.org/10.1109/ICDMW.2009.94

@inproceedings{c3128f52b04d4ed5af04b7796ad701c2,

title = "Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data",

abstract = "The paper focuses on automated categorization of text documents, each labeled with one or more classes and described by tens of thousands of features. The computational costs of induction in such domains are so high as almost to disqualify the use of decision trees; the reduction of these costs is thus an important research issue. Our own solution, FDT ({"}fast decision-tree induction{"}), uses a two-pronged strategy: (1) feature-set pre-selection, and (2) induction of several trees, each from a different data subset, with the combination of the results from multiple trees with a data-fusion technique tailored to domains with imbalanced classes.",

keywords = "Decision tree, Imbalanced classes, Large-scale data, Multi-label examples, Text categorization",

author = "Peerapon Vateekul and Miroslav Kubat",

year = "2009",

month = dec,

day = "1",

doi = "10.1109/ICDMW.2009.94",

language = "English (US)",

isbn = "9780769539027",

series = "ICDM Workshops 2009 - IEEE International Conference on Data Mining",

pages = "320--325",

booktitle = "ICDM Workshops 2009 - IEEE International Conference on Data Mining",

note = "2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009 ; Conference date: 06-12-2009 Through 06-12-2009",

}

TY - GEN

T1 - Fast induction of multiple decision trees in text categorization from large scale, imbalanced, and multi-label data

AU - Vateekul, Peerapon

AU - Kubat, Miroslav

PY - 2009/12/1

Y1 - 2009/12/1

N2 - The paper focuses on automated categorization of text documents, each labeled with one or more classes and described by tens of thousands of features. The computational costs of induction in such domains are so high as almost to disqualify the use of decision trees; the reduction of these costs is thus an important research issue. Our own solution, FDT ("fast decision-tree induction"), uses a two-pronged strategy: (1) feature-set pre-selection, and (2) induction of several trees, each from a different data subset, with the combination of the results from multiple trees with a data-fusion technique tailored to domains with imbalanced classes.

AB - The paper focuses on automated categorization of text documents, each labeled with one or more classes and described by tens of thousands of features. The computational costs of induction in such domains are so high as almost to disqualify the use of decision trees; the reduction of these costs is thus an important research issue. Our own solution, FDT ("fast decision-tree induction"), uses a two-pronged strategy: (1) feature-set pre-selection, and (2) induction of several trees, each from a different data subset, with the combination of the results from multiple trees with a data-fusion technique tailored to domains with imbalanced classes.

KW - Decision tree

KW - Imbalanced classes

KW - Large-scale data

KW - Multi-label examples

KW - Text categorization

UR - http://www.scopus.com/inward/record.url?scp=77951150128&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951150128&partnerID=8YFLogxK

U2 - 10.1109/ICDMW.2009.94

DO - 10.1109/ICDMW.2009.94

M3 - Conference contribution

AN - SCOPUS:77951150128

SN - 9780769539027

T3 - ICDM Workshops 2009 - IEEE International Conference on Data Mining

SP - 320

EP - 325

BT - ICDM Workshops 2009 - IEEE International Conference on Data Mining

T2 - 2009 IEEE International Conference on Data Mining Workshops, ICDMW 2009

Y2 - 6 December 2009 through 6 December 2009

ER -