Evolving data sets to highlight the performance differences between machine learning classifiers

Thomas Raway, J. David Schaffer, Kenneth J. Kurtz, Hiroki Sayama

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We present a preliminary study to evolve data sets that maximize performance differences between multiple machine learning classifiers. The aim is to provide useful information towards the decision of which machine learning classifier to use given a particular data set. While literature already exists on comparing multiple classifiers across multiple pre-existing data sets, our approach is novel and unique in that we evolved completely new data sets designed to highlight the performance differences between supervised learning classifiers. By investigating these evolved data sets, we hope to add to the knowledge base concerning which classifiers are appropriate for specific real world classification tasks. Copyright is held by the author/owner(s).

Original languageEnglish
Title of host publicationGECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion
Pages657-658
Number of pages2
DOIs
Publication statusPublished - 2012
Externally publishedYes
Event14th International Conference on Genetic and Evolutionary Computation, GECCO'12 - Philadelphia, PA, United States
Duration: 2012 Jul 72012 Jul 11

Other

Other14th International Conference on Genetic and Evolutionary Computation, GECCO'12
CountryUnited States
CityPhiladelphia, PA
Period12/7/712/7/11

Fingerprint

Learning systems
Classifiers
Supervised learning

Keywords

  • Complexity measures
  • Evolutionary computation
  • Machine learning

ASJC Scopus subject areas

  • Computational Theory and Mathematics

Cite this

Raway, T., Schaffer, J. D., Kurtz, K. J., & Sayama, H. (2012). Evolving data sets to highlight the performance differences between machine learning classifiers. In GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion (pp. 657-658) https://doi.org/10.1145/2330784.2330907

Evolving data sets to highlight the performance differences between machine learning classifiers. / Raway, Thomas; Schaffer, J. David; Kurtz, Kenneth J.; Sayama, Hiroki.

GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion. 2012. p. 657-658.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Raway, T, Schaffer, JD, Kurtz, KJ & Sayama, H 2012, Evolving data sets to highlight the performance differences between machine learning classifiers. in GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion. pp. 657-658, 14th International Conference on Genetic and Evolutionary Computation, GECCO'12, Philadelphia, PA, United States, 12/7/7. https://doi.org/10.1145/2330784.2330907
Raway T, Schaffer JD, Kurtz KJ, Sayama H. Evolving data sets to highlight the performance differences between machine learning classifiers. In GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion. 2012. p. 657-658 https://doi.org/10.1145/2330784.2330907
Raway, Thomas ; Schaffer, J. David ; Kurtz, Kenneth J. ; Sayama, Hiroki. / Evolving data sets to highlight the performance differences between machine learning classifiers. GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion. 2012. pp. 657-658
@inproceedings{4d78bc15aa9b480a942cf5be52a02b52,
title = "Evolving data sets to highlight the performance differences between machine learning classifiers",
abstract = "We present a preliminary study to evolve data sets that maximize performance differences between multiple machine learning classifiers. The aim is to provide useful information towards the decision of which machine learning classifier to use given a particular data set. While literature already exists on comparing multiple classifiers across multiple pre-existing data sets, our approach is novel and unique in that we evolved completely new data sets designed to highlight the performance differences between supervised learning classifiers. By investigating these evolved data sets, we hope to add to the knowledge base concerning which classifiers are appropriate for specific real world classification tasks. Copyright is held by the author/owner(s).",
keywords = "Complexity measures, Evolutionary computation, Machine learning",
author = "Thomas Raway and Schaffer, {J. David} and Kurtz, {Kenneth J.} and Hiroki Sayama",
year = "2012",
doi = "10.1145/2330784.2330907",
language = "English",
isbn = "9781450311786",
pages = "657--658",
booktitle = "GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion",

}

TY - GEN

T1 - Evolving data sets to highlight the performance differences between machine learning classifiers

AU - Raway, Thomas

AU - Schaffer, J. David

AU - Kurtz, Kenneth J.

AU - Sayama, Hiroki

PY - 2012

Y1 - 2012

N2 - We present a preliminary study to evolve data sets that maximize performance differences between multiple machine learning classifiers. The aim is to provide useful information towards the decision of which machine learning classifier to use given a particular data set. While literature already exists on comparing multiple classifiers across multiple pre-existing data sets, our approach is novel and unique in that we evolved completely new data sets designed to highlight the performance differences between supervised learning classifiers. By investigating these evolved data sets, we hope to add to the knowledge base concerning which classifiers are appropriate for specific real world classification tasks. Copyright is held by the author/owner(s).

AB - We present a preliminary study to evolve data sets that maximize performance differences between multiple machine learning classifiers. The aim is to provide useful information towards the decision of which machine learning classifier to use given a particular data set. While literature already exists on comparing multiple classifiers across multiple pre-existing data sets, our approach is novel and unique in that we evolved completely new data sets designed to highlight the performance differences between supervised learning classifiers. By investigating these evolved data sets, we hope to add to the knowledge base concerning which classifiers are appropriate for specific real world classification tasks. Copyright is held by the author/owner(s).

KW - Complexity measures

KW - Evolutionary computation

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=84864985846&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864985846&partnerID=8YFLogxK

U2 - 10.1145/2330784.2330907

DO - 10.1145/2330784.2330907

M3 - Conference contribution

SN - 9781450311786

SP - 657

EP - 658

BT - GECCO'12 - Proceedings of the 14th International Conference on Genetic and Evolutionary Computation Companion

ER -