A constructive meta-level feature selection method based on method repositories

Hidenao Abe, Takahira Yamaguchi

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Feature selection is one of key issues related with data pre-processing of classification task in a data mining process. Although many efforts have been done to improve typical feature selection algorithms (FSAs), such as filter methods and wrapper methods, it is hard for just one FSA to manage its performances to various datasets. To above problems, we propose another way to support feature selection procedure, constructing proper FSAs to each given dataset. Here is discussed constructive metalevel feature selection that re-constructs proper FSAs with a method repository every given datasets, de-composing representative FSAs into methods. After implementing the constructive meta-level feature selection system, we show how constructive meta-level feature selection goes well with 34 UCI common data sets, comparing with typical FSAs on their accuracies. As the result, our system shows the high performance on accuracies with lower computational costs to construct a proper FSA to each given data set automatically.

Original languageEnglish
Pages (from-to)20-26
Number of pages7
JournalJournal of Computers
Volume1
Issue number3
Publication statusPublished - 2006

Fingerprint

Feature extraction
Data mining

Keywords

  • Constructive Meta-Processing
  • Data Mining
  • Feature Selection

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

A constructive meta-level feature selection method based on method repositories. / Abe, Hidenao; Yamaguchi, Takahira.

In: Journal of Computers, Vol. 1, No. 3, 2006, p. 20-26.

Research output: Contribution to journalArticle

@article{493bec76a29a4e88affadd000cfccaa4,
title = "A constructive meta-level feature selection method based on method repositories",
abstract = "Feature selection is one of key issues related with data pre-processing of classification task in a data mining process. Although many efforts have been done to improve typical feature selection algorithms (FSAs), such as filter methods and wrapper methods, it is hard for just one FSA to manage its performances to various datasets. To above problems, we propose another way to support feature selection procedure, constructing proper FSAs to each given dataset. Here is discussed constructive metalevel feature selection that re-constructs proper FSAs with a method repository every given datasets, de-composing representative FSAs into methods. After implementing the constructive meta-level feature selection system, we show how constructive meta-level feature selection goes well with 34 UCI common data sets, comparing with typical FSAs on their accuracies. As the result, our system shows the high performance on accuracies with lower computational costs to construct a proper FSA to each given data set automatically.",
keywords = "Constructive Meta-Processing, Data Mining, Feature Selection",
author = "Hidenao Abe and Takahira Yamaguchi",
year = "2006",
language = "English",
volume = "1",
pages = "20--26",
journal = "Journal of Computers",
issn = "1796-203X",
publisher = "Academy Publisher",
number = "3",

}

TY - JOUR

T1 - A constructive meta-level feature selection method based on method repositories

AU - Abe, Hidenao

AU - Yamaguchi, Takahira

PY - 2006

Y1 - 2006

N2 - Feature selection is one of key issues related with data pre-processing of classification task in a data mining process. Although many efforts have been done to improve typical feature selection algorithms (FSAs), such as filter methods and wrapper methods, it is hard for just one FSA to manage its performances to various datasets. To above problems, we propose another way to support feature selection procedure, constructing proper FSAs to each given dataset. Here is discussed constructive metalevel feature selection that re-constructs proper FSAs with a method repository every given datasets, de-composing representative FSAs into methods. After implementing the constructive meta-level feature selection system, we show how constructive meta-level feature selection goes well with 34 UCI common data sets, comparing with typical FSAs on their accuracies. As the result, our system shows the high performance on accuracies with lower computational costs to construct a proper FSA to each given data set automatically.

AB - Feature selection is one of key issues related with data pre-processing of classification task in a data mining process. Although many efforts have been done to improve typical feature selection algorithms (FSAs), such as filter methods and wrapper methods, it is hard for just one FSA to manage its performances to various datasets. To above problems, we propose another way to support feature selection procedure, constructing proper FSAs to each given dataset. Here is discussed constructive metalevel feature selection that re-constructs proper FSAs with a method repository every given datasets, de-composing representative FSAs into methods. After implementing the constructive meta-level feature selection system, we show how constructive meta-level feature selection goes well with 34 UCI common data sets, comparing with typical FSAs on their accuracies. As the result, our system shows the high performance on accuracies with lower computational costs to construct a proper FSA to each given data set automatically.

KW - Constructive Meta-Processing

KW - Data Mining

KW - Feature Selection

UR - http://www.scopus.com/inward/record.url?scp=84866647433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866647433&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84866647433

VL - 1

SP - 20

EP - 26

JO - Journal of Computers

JF - Journal of Computers

SN - 1796-203X

IS - 3

ER -