IPRed: Instance reduction algorithm based on the percentile of the partitions

Turki Turki; Zhi Wei

IPRed: Instance reduction algorithm based on the percentile of the partitions

Turki Turki, Zhi Wei

Research output: Contribution to journal › Conference article › peer-review

Abstract

Instance reduction methods are popular methods that reduce the size of the datasets to possibly improve the classification accuracy. We present a method that reduces the size of the dataset based on the percentile of the dataset partitions which we call IPRed. We evaluate our and other popular instance reduction methods from a classification perspective by 1-nearest neighbor algorithm on many real datasets. Our experimental evaluation on the datasets shows that our method yields the minimum average error with statistical significance.

Original language	English (US)
Pages (from-to)	181-185
Number of pages	5
Journal	CEUR Workshop Proceedings
Volume	1353
State	Published - 2015
Externally published	Yes
Event	26th Modern Artificial Intelligence and Cognitive Science Conference, MAICS 2015 - Greensboro, United States Duration: Apr 25 2015 → Apr 26 2015

All Science Journal Classification (ASJC) codes

General Computer Science

Keywords

Classification
Instance reduction
Nearest neighbor
Statistical significance

Cite this

@article{a8bcfb1ef531406a8d7b675eb429764d,

title = "IPRed: Instance reduction algorithm based on the percentile of the partitions",

abstract = "Instance reduction methods are popular methods that reduce the size of the datasets to possibly improve the classification accuracy. We present a method that reduces the size of the dataset based on the percentile of the dataset partitions which we call IPRed. We evaluate our and other popular instance reduction methods from a classification perspective by 1-nearest neighbor algorithm on many real datasets. Our experimental evaluation on the datasets shows that our method yields the minimum average error with statistical significance.",

keywords = "Classification, Instance reduction, Nearest neighbor, Statistical significance",

author = "Turki Turki and Zhi Wei",

year = "2015",

language = "English (US)",

volume = "1353",

pages = "181--185",

journal = "CEUR Workshop Proceedings",

issn = "1613-0073",

publisher = "CEUR-WS",

note = "26th Modern Artificial Intelligence and Cognitive Science Conference, MAICS 2015 ; Conference date: 25-04-2015 Through 26-04-2015",

}

TY - JOUR

T1 - IPRed

T2 - 26th Modern Artificial Intelligence and Cognitive Science Conference, MAICS 2015

AU - Turki, Turki

AU - Wei, Zhi

PY - 2015

Y1 - 2015

N2 - Instance reduction methods are popular methods that reduce the size of the datasets to possibly improve the classification accuracy. We present a method that reduces the size of the dataset based on the percentile of the dataset partitions which we call IPRed. We evaluate our and other popular instance reduction methods from a classification perspective by 1-nearest neighbor algorithm on many real datasets. Our experimental evaluation on the datasets shows that our method yields the minimum average error with statistical significance.

AB - Instance reduction methods are popular methods that reduce the size of the datasets to possibly improve the classification accuracy. We present a method that reduces the size of the dataset based on the percentile of the dataset partitions which we call IPRed. We evaluate our and other popular instance reduction methods from a classification perspective by 1-nearest neighbor algorithm on many real datasets. Our experimental evaluation on the datasets shows that our method yields the minimum average error with statistical significance.

KW - Classification

KW - Instance reduction

KW - Nearest neighbor

KW - Statistical significance

UR - http://www.scopus.com/inward/record.url?scp=84928711886&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84928711886&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:84928711886

SN - 1613-0073

VL - 1353

SP - 181

EP - 185

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

Y2 - 25 April 2015 through 26 April 2015

ER -

IPRed: Instance reduction algorithm based on the percentile of the partitions

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Other files and links

Fingerprint

Cite this