A balanced ensemble approach to weighting classifiers for text classification

Gabriel Pui Cheong Fung, Jeffrey Xu Yu, Haixun Wang, David W. Cheung, Huan Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Citations (Scopus)

Abstract

This paper studies the problem of constructing an effective heterogeneous ensemble classifier for text classification. One major challenge of this problem is to formulate a good combination function, which combines the decisions of the individual classifiers in the ensemble. We show that the classification performance is affected by three weight components and they should be included in deriving an effective combination function. They are: (1) Global effectiveness, which measures the effectiveness of a member classifier in classifying a set of unseen documents; (2) Local effectiveness, which measures the effectiveness of a member classifier in classifying the particular domain of an unseen document; and (3) Decision confidence, which describes how confident a classifier is when making a decision when classifying a specific unseen document. We propose a new balanced combination function, called Dynamic Classifier Weighting (DCW), that incorporates the aforementioned three components. The empirical study demonstrates that the new combination function is highly effective for text classification.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE International Conference on Data Mining, ICDM
Pages869-873
Number of pages5
DOIs
StatePublished - 2006
Event6th International Conference on Data Mining, ICDM 2006 - Hong Kong, China
Duration: Dec 18 2006Dec 22 2006

Other

Other6th International Conference on Data Mining, ICDM 2006
CountryChina
CityHong Kong
Period12/18/0612/22/06

Fingerprint

Classifiers

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Fung, G. P. C., Yu, J. X., Wang, H., Cheung, D. W., & Liu, H. (2006). A balanced ensemble approach to weighting classifiers for text classification. In Proceedings - IEEE International Conference on Data Mining, ICDM (pp. 869-873). [4053118] https://doi.org/10.1109/ICDM.2006.2

A balanced ensemble approach to weighting classifiers for text classification. / Fung, Gabriel Pui Cheong; Yu, Jeffrey Xu; Wang, Haixun; Cheung, David W.; Liu, Huan.

Proceedings - IEEE International Conference on Data Mining, ICDM. 2006. p. 869-873 4053118.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Fung, GPC, Yu, JX, Wang, H, Cheung, DW & Liu, H 2006, A balanced ensemble approach to weighting classifiers for text classification. in Proceedings - IEEE International Conference on Data Mining, ICDM., 4053118, pp. 869-873, 6th International Conference on Data Mining, ICDM 2006, Hong Kong, China, 12/18/06. https://doi.org/10.1109/ICDM.2006.2
Fung GPC, Yu JX, Wang H, Cheung DW, Liu H. A balanced ensemble approach to weighting classifiers for text classification. In Proceedings - IEEE International Conference on Data Mining, ICDM. 2006. p. 869-873. 4053118 https://doi.org/10.1109/ICDM.2006.2
Fung, Gabriel Pui Cheong ; Yu, Jeffrey Xu ; Wang, Haixun ; Cheung, David W. ; Liu, Huan. / A balanced ensemble approach to weighting classifiers for text classification. Proceedings - IEEE International Conference on Data Mining, ICDM. 2006. pp. 869-873
@inproceedings{f2c08056464c4b35abcda3822a13405b,
title = "A balanced ensemble approach to weighting classifiers for text classification",
abstract = "This paper studies the problem of constructing an effective heterogeneous ensemble classifier for text classification. One major challenge of this problem is to formulate a good combination function, which combines the decisions of the individual classifiers in the ensemble. We show that the classification performance is affected by three weight components and they should be included in deriving an effective combination function. They are: (1) Global effectiveness, which measures the effectiveness of a member classifier in classifying a set of unseen documents; (2) Local effectiveness, which measures the effectiveness of a member classifier in classifying the particular domain of an unseen document; and (3) Decision confidence, which describes how confident a classifier is when making a decision when classifying a specific unseen document. We propose a new balanced combination function, called Dynamic Classifier Weighting (DCW), that incorporates the aforementioned three components. The empirical study demonstrates that the new combination function is highly effective for text classification.",
author = "Fung, {Gabriel Pui Cheong} and Yu, {Jeffrey Xu} and Haixun Wang and Cheung, {David W.} and Huan Liu",
year = "2006",
doi = "10.1109/ICDM.2006.2",
language = "English (US)",
isbn = "0769527019",
pages = "869--873",
booktitle = "Proceedings - IEEE International Conference on Data Mining, ICDM",

}

TY - GEN

T1 - A balanced ensemble approach to weighting classifiers for text classification

AU - Fung, Gabriel Pui Cheong

AU - Yu, Jeffrey Xu

AU - Wang, Haixun

AU - Cheung, David W.

AU - Liu, Huan

PY - 2006

Y1 - 2006

N2 - This paper studies the problem of constructing an effective heterogeneous ensemble classifier for text classification. One major challenge of this problem is to formulate a good combination function, which combines the decisions of the individual classifiers in the ensemble. We show that the classification performance is affected by three weight components and they should be included in deriving an effective combination function. They are: (1) Global effectiveness, which measures the effectiveness of a member classifier in classifying a set of unseen documents; (2) Local effectiveness, which measures the effectiveness of a member classifier in classifying the particular domain of an unseen document; and (3) Decision confidence, which describes how confident a classifier is when making a decision when classifying a specific unseen document. We propose a new balanced combination function, called Dynamic Classifier Weighting (DCW), that incorporates the aforementioned three components. The empirical study demonstrates that the new combination function is highly effective for text classification.

AB - This paper studies the problem of constructing an effective heterogeneous ensemble classifier for text classification. One major challenge of this problem is to formulate a good combination function, which combines the decisions of the individual classifiers in the ensemble. We show that the classification performance is affected by three weight components and they should be included in deriving an effective combination function. They are: (1) Global effectiveness, which measures the effectiveness of a member classifier in classifying a set of unseen documents; (2) Local effectiveness, which measures the effectiveness of a member classifier in classifying the particular domain of an unseen document; and (3) Decision confidence, which describes how confident a classifier is when making a decision when classifying a specific unseen document. We propose a new balanced combination function, called Dynamic Classifier Weighting (DCW), that incorporates the aforementioned three components. The empirical study demonstrates that the new combination function is highly effective for text classification.

UR - http://www.scopus.com/inward/record.url?scp=84878065641&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878065641&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2006.2

DO - 10.1109/ICDM.2006.2

M3 - Conference contribution

AN - SCOPUS:84878065641

SN - 0769527019

SN - 9780769527017

SP - 869

EP - 873

BT - Proceedings - IEEE International Conference on Data Mining, ICDM

ER -