Meta-learning for data summarization based on instance selection method

Kate Smith-Miles; Rafiqul Islam

doi:10.1109/CEC.2010.5585986

Meta-learning for data summarization based on instance selection method

Kate Smith-Miles, Rafiqul Islam

School of Mathematics

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

11 Citations (Scopus)

Abstract

The purpose of instance selection is to identify which instances (examples, patterns) in a large dataset should be selected as representatives of the entire dataset, without significant loss of information. When a machine learning method is applied to the reduced dataset, the accuracy of the model should not be significantly worse than if the same method were applied to the entire dataset. The reducibility of any dataset, and hence the success of instance selection methods, surely depends on the characteristics of the dataset, as well as the machine learning method. This paper adopts a meta-learning approach, via an empirical study of 112 classification datasets from the UCI Repository, to explore the relationship between data characteristics, machine learning methods, and the success of instance selection method.

Original language	English
Title of host publication	Proceedings of the 2010 IEEE Congress on Evolutionary Computation (CEC),
Editors	Gary Fogel
Place of Publication	Atlanta USA
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1 - 8
Number of pages	8
DOIs	https://doi.org/10.1109/CEC.2010.5585986
Publication status	Published - 2010
Event	IEEE Congress on Evolutionary Computation 2010 - Barcelona, Spain Duration: 18 Jul 2010 → 23 Jul 2010 https://ieeexplore.ieee.org/xpl/conhome/5573635/proceeding (Proceedings)

Conference

Conference	IEEE Congress on Evolutionary Computation 2010
Abbreviated title	IEEE CEC 2010
Country/Territory	Spain
City	Barcelona
Period	18/07/10 → 23/07/10
Internet address	https://ieeexplore.ieee.org/xpl/conhome/5573635/proceeding (Proceedings)

Access to Document

10.1109/CEC.2010.5585986

Cite this

@inproceedings{2401813b3734464ab67621b69fe377c5,

title = "Meta-learning for data summarization based on instance selection method",

abstract = "The purpose of instance selection is to identify which instances (examples, patterns) in a large dataset should be selected as representatives of the entire dataset, without significant loss of information. When a machine learning method is applied to the reduced dataset, the accuracy of the model should not be significantly worse than if the same method were applied to the entire dataset. The reducibility of any dataset, and hence the success of instance selection methods, surely depends on the characteristics of the dataset, as well as the machine learning method. This paper adopts a meta-learning approach, via an empirical study of 112 classification datasets from the UCI Repository, to explore the relationship between data characteristics, machine learning methods, and the success of instance selection method.",

author = "Kate Smith-Miles and Rafiqul Islam",

year = "2010",

doi = "10.1109/CEC.2010.5585986",

language = "English",

pages = "1 -- 8",

editor = "Gary Fogel",

booktitle = "Proceedings of the 2010 IEEE Congress on Evolutionary Computation (CEC),",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

note = "IEEE Congress on Evolutionary Computation 2010, IEEE CEC 2010 ; Conference date: 18-07-2010 Through 23-07-2010",

url = "https://ieeexplore.ieee.org/xpl/conhome/5573635/proceeding",

}

Smith-Miles, K & Islam, R 2010, Meta-learning for data summarization based on instance selection method. in G Fogel (ed.), Proceedings of the 2010 IEEE Congress on Evolutionary Computation (CEC),. IEEE, Institute of Electrical and Electronics Engineers, Atlanta USA, pp. 1 - 8, IEEE Congress on Evolutionary Computation 2010, Barcelona, Spain, 18/07/10. https://doi.org/10.1109/CEC.2010.5585986

Meta-learning for data summarization based on instance selection method. / Smith-Miles, Kate; Islam, Rafiqul.
Proceedings of the 2010 IEEE Congress on Evolutionary Computation (CEC),. ed. / Gary Fogel. Atlanta USA: IEEE, Institute of Electrical and Electronics Engineers, 2010. p. 1 - 8.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Meta-learning for data summarization based on instance selection method

AU - Smith-Miles, Kate

AU - Islam, Rafiqul

PY - 2010

Y1 - 2010

N2 - The purpose of instance selection is to identify which instances (examples, patterns) in a large dataset should be selected as representatives of the entire dataset, without significant loss of information. When a machine learning method is applied to the reduced dataset, the accuracy of the model should not be significantly worse than if the same method were applied to the entire dataset. The reducibility of any dataset, and hence the success of instance selection methods, surely depends on the characteristics of the dataset, as well as the machine learning method. This paper adopts a meta-learning approach, via an empirical study of 112 classification datasets from the UCI Repository, to explore the relationship between data characteristics, machine learning methods, and the success of instance selection method.

AB - The purpose of instance selection is to identify which instances (examples, patterns) in a large dataset should be selected as representatives of the entire dataset, without significant loss of information. When a machine learning method is applied to the reduced dataset, the accuracy of the model should not be significantly worse than if the same method were applied to the entire dataset. The reducibility of any dataset, and hence the success of instance selection methods, surely depends on the characteristics of the dataset, as well as the machine learning method. This paper adopts a meta-learning approach, via an empirical study of 112 classification datasets from the UCI Repository, to explore the relationship between data characteristics, machine learning methods, and the success of instance selection method.

UR - http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5585986

U2 - 10.1109/CEC.2010.5585986

DO - 10.1109/CEC.2010.5585986

M3 - Conference Paper

SP - 1

EP - 8

BT - Proceedings of the 2010 IEEE Congress on Evolutionary Computation (CEC),

A2 - Fogel, Gary

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Atlanta USA

T2 - IEEE Congress on Evolutionary Computation 2010

Y2 - 18 July 2010 through 23 July 2010

ER -

Meta-learning for data summarization based on instance selection method

Abstract

Conference

Access to Document

Other files and links

Cite this