Attribute Selection Methods for Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR)

IM Whittley; AJ Bagnall; L Bull; M Pettipher; M Studley; F Tekiner

Attribute Selection Methods for Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR)

IM Whittley, AJ Bagnall, L Bull, M Pettipher, M Studley, F Tekiner

Research output: Contribution to conference › Paper

Abstract

Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR) is a recently proposed algorithm for ensembles of k-nn classifiers [28]. FASBIR works by first performing a global filtering of attributes using information gain, then randomising the bagged ensemble with random subsets of the remaining attributes and random distance metrics. In this paper we propose two refinements of FASBIR and evaluate them on several very large data sets.

Original language	English
Pages	10-17
Number of pages	8
Publication status	Published - 2006
Event	Feature Selection for Data Mining Workshop, Part of the 2006 SIAM Conference on Data Mining - Bethesda, United States Duration: 22 Apr 2006 → …

Conference

Conference	Feature Selection for Data Mining Workshop, Part of the 2006 SIAM Conference on Data Mining
Country/Territory	United States
City	Bethesda
Period	22/04/06 → …

Cite this

@conference{fad6348b01ec435eabc05b9d042c0269,

title = "Attribute Selection Methods for Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR)",

abstract = "Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR) is a recently proposed algorithm for ensembles of k-nn classifiers [28]. FASBIR works by first performing a global filtering of attributes using information gain, then randomising the bagged ensemble with random subsets of the remaining attributes and random distance metrics. In this paper we propose two refinements of FASBIR and evaluate them on several very large data sets.",

author = "IM Whittley and AJ Bagnall and L Bull and M Pettipher and M Studley and F Tekiner",

year = "2006",

language = "English",

pages = "10--17",

note = "Feature Selection for Data Mining Workshop, Part of the 2006 SIAM Conference on Data Mining ; Conference date: 22-04-2006",

}

TY - CONF

T1 - Attribute Selection Methods for Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR)

AU - Whittley, IM

AU - Bagnall, AJ

AU - Bull, L

AU - Pettipher, M

AU - Studley, M

AU - Tekiner, F

PY - 2006

Y1 - 2006

N2 - Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR) is a recently proposed algorithm for ensembles of k-nn classifiers [28]. FASBIR works by first performing a global filtering of attributes using information gain, then randomising the bagged ensemble with random subsets of the remaining attributes and random distance metrics. In this paper we propose two refinements of FASBIR and evaluate them on several very large data sets.

AB - Filtered Attribute Subspace based Bagging with Injected Randomness (FASBIR) is a recently proposed algorithm for ensembles of k-nn classifiers [28]. FASBIR works by first performing a global filtering of attributes using information gain, then randomising the bagged ensemble with random subsets of the remaining attributes and random distance metrics. In this paper we propose two refinements of FASBIR and evaluate them on several very large data sets.

M3 - Paper

SP - 10

EP - 17

T2 - Feature Selection for Data Mining Workshop, Part of the 2006 SIAM Conference on Data Mining

Y2 - 22 April 2006

ER -