WBC_SVM: Weighted Bayesian Classification based on Support  Vector Machines

T Gaertner; PA Flach

WBC_SVM: Weighted Bayesian Classification based on Support Vector Machines

T Gaertner, PA Flach

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

Abstract

This paper introduces an algorithm that combines nay?ve Bayes classification with feature weighting. Most of the related approaches to feature transformation for nay?ve Bayes suggest various heuristics and non-exhaustive search strategies for selecting a subset of features with which nay?ve Bayes performs better than with the complete set of features. In contrast, the algorithm introduced in this paper employs feature weighting performed by a support vector machine. The weights are optimised such that the danger of overfitting is reduced. To the best of our knowledge, this is the first time that nay?ve Bayes classification has been combined with feature weighting. Experimental results on 15 UCI domains demonstrate that WBCSVM compares favourably to state-of-the-art machine learning approaches.

Translated title of the contribution	WBC_SVM: Weighted Bayesian Classification based on Support Vector Machines
Original language	English
Title of host publication	Unknown
Editors	Carla E. Brodley, Andrea Pohoreckyj Danyluk
Publisher	Morgan Kaufmann
Pages	207 - 209
Number of pages	2
Publication status	Published - Jun 2001

Bibliographical note

Conference Proceedings/Title of Journal: Proceedings of the Eighteenth International Conference on Machine Learning

Access to Document

http://www.cs.bris.ac.uk/Publications/pub_info.jsp?id=1000560

Handle.net

Persistent link

Cite this

@inproceedings{5e3956a979184537a588aec50343dfc6,

title = "WBC_SVM: Weighted Bayesian Classification based on Support Vector Machines",

abstract = "This paper introduces an algorithm that combines nay?ve Bayes classification with feature weighting. Most of the related approaches to feature transformation for nay?ve Bayes suggest various heuristics and non-exhaustive search strategies for selecting a subset of features with which nay?ve Bayes performs better than with the complete set of features. In contrast, the algorithm introduced in this paper employs feature weighting performed by a support vector machine. The weights are optimised such that the danger of overfitting is reduced. To the best of our knowledge, this is the first time that nay?ve Bayes classification has been combined with feature weighting. Experimental results on 15 UCI domains demonstrate that WBCSVM compares favourably to state-of-the-art machine learning approaches. ",

author = "T Gaertner and PA Flach",

note = "Conference Proceedings/Title of Journal: Proceedings of the Eighteenth International Conference on Machine Learning",

year = "2001",

month = jun,

language = "English",

pages = "207 -- 209",

editor = "Brodley, {Carla E.} and Danyluk, {Andrea Pohoreckyj}",

booktitle = "Unknown",

publisher = "Morgan Kaufmann",

}

TY - GEN

T1 - WBC_SVM: Weighted Bayesian Classification based on Support Vector Machines

AU - Gaertner, T

AU - Flach, PA

N1 - Conference Proceedings/Title of Journal: Proceedings of the Eighteenth International Conference on Machine Learning

PY - 2001/6

Y1 - 2001/6

N2 - This paper introduces an algorithm that combines nay?ve Bayes classification with feature weighting. Most of the related approaches to feature transformation for nay?ve Bayes suggest various heuristics and non-exhaustive search strategies for selecting a subset of features with which nay?ve Bayes performs better than with the complete set of features. In contrast, the algorithm introduced in this paper employs feature weighting performed by a support vector machine. The weights are optimised such that the danger of overfitting is reduced. To the best of our knowledge, this is the first time that nay?ve Bayes classification has been combined with feature weighting. Experimental results on 15 UCI domains demonstrate that WBCSVM compares favourably to state-of-the-art machine learning approaches.

AB - This paper introduces an algorithm that combines nay?ve Bayes classification with feature weighting. Most of the related approaches to feature transformation for nay?ve Bayes suggest various heuristics and non-exhaustive search strategies for selecting a subset of features with which nay?ve Bayes performs better than with the complete set of features. In contrast, the algorithm introduced in this paper employs feature weighting performed by a support vector machine. The weights are optimised such that the danger of overfitting is reduced. To the best of our knowledge, this is the first time that nay?ve Bayes classification has been combined with feature weighting. Experimental results on 15 UCI domains demonstrate that WBCSVM compares favourably to state-of-the-art machine learning approaches.

M3 - Conference Contribution (Conference Proceeding)

SP - 207

EP - 209

BT - Unknown

A2 - Brodley, Carla E.

A2 - Danyluk, Andrea Pohoreckyj

PB - Morgan Kaufmann