A comparative study of several genetic-based supervised learning systems

Albert Orriols-Puig; Jorge Casillas; Ester Bernadó-Mansilla

doi:10.1007/978-3-540-78979-6_10

A comparative study of several genetic-based supervised learning systems

Albert Orriols-Puig^*, Jorge Casillas, Ester Bernadó-Mansilla

^*Corresponding author for this work

Research output: Book chapter › Chapter › peer-review

6 Citations (Scopus)

Abstract

This chapter gives insight in the use of Genetic-Based Machine Learning (GBML) for supervised tasks. Five GBML systems which represent different learning methodologies and knowledge representations in the GBML paradigm are selected for the analysis: UCS, GAssist, SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost. UCS and GAssist are based on a non-fuzzy representation, while SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost use a linguistic fuzzy representation. The models evolved by these five systems are compared in terms of performance and interpretability to the models created by six highly-used non-evolutionary learners. Experimental observations highlight the suitability of GBML systems for classification tasks. Moreover, the analysis points out which systems should be used depending on whether the user prefers to maximize the accuracy or the interpretability of the models.

Original language	English
Title of host publication	Learning Classifier Systems in Data Mining
Editors	Larry Bull, Ester Bernadó-Mansilla, John Holmes
Pages	205-230
Number of pages	26
DOIs	https://doi.org/10.1007/978-3-540-78979-6_10
Publication status	Published - 2008

Publication series

Name	Studies in Computational Intelligence
Volume	125
ISSN (Print)	1860-949X

Access to Document

10.1007/978-3-540-78979-6_10

Cite this

@inbook{191117a958e44b9fab5227514a6eff73,

title = "A comparative study of several genetic-based supervised learning systems",

abstract = "This chapter gives insight in the use of Genetic-Based Machine Learning (GBML) for supervised tasks. Five GBML systems which represent different learning methodologies and knowledge representations in the GBML paradigm are selected for the analysis: UCS, GAssist, SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost. UCS and GAssist are based on a non-fuzzy representation, while SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost use a linguistic fuzzy representation. The models evolved by these five systems are compared in terms of performance and interpretability to the models created by six highly-used non-evolutionary learners. Experimental observations highlight the suitability of GBML systems for classification tasks. Moreover, the analysis points out which systems should be used depending on whether the user prefers to maximize the accuracy or the interpretability of the models.",

author = "Albert Orriols-Puig and Jorge Casillas and Ester Bernad{\'o}-Mansilla",

year = "2008",

doi = "10.1007/978-3-540-78979-6_10",

language = "English",

isbn = "9783540789789",

series = "Studies in Computational Intelligence",

pages = "205--230",

editor = "Larry Bull and Ester Bernad{\'o}-Mansilla and John Holmes",

booktitle = "Learning Classifier Systems in Data Mining",

}

TY - CHAP

T1 - A comparative study of several genetic-based supervised learning systems

AU - Orriols-Puig, Albert

AU - Casillas, Jorge

AU - Bernadó-Mansilla, Ester

PY - 2008

Y1 - 2008

N2 - This chapter gives insight in the use of Genetic-Based Machine Learning (GBML) for supervised tasks. Five GBML systems which represent different learning methodologies and knowledge representations in the GBML paradigm are selected for the analysis: UCS, GAssist, SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost. UCS and GAssist are based on a non-fuzzy representation, while SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost use a linguistic fuzzy representation. The models evolved by these five systems are compared in terms of performance and interpretability to the models created by six highly-used non-evolutionary learners. Experimental observations highlight the suitability of GBML systems for classification tasks. Moreover, the analysis points out which systems should be used depending on whether the user prefers to maximize the accuracy or the interpretability of the models.

AB - This chapter gives insight in the use of Genetic-Based Machine Learning (GBML) for supervised tasks. Five GBML systems which represent different learning methodologies and knowledge representations in the GBML paradigm are selected for the analysis: UCS, GAssist, SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost. UCS and GAssist are based on a non-fuzzy representation, while SLAVE, Fuzzy AdaBoost, and Fuzzy LogitBoost use a linguistic fuzzy representation. The models evolved by these five systems are compared in terms of performance and interpretability to the models created by six highly-used non-evolutionary learners. Experimental observations highlight the suitability of GBML systems for classification tasks. Moreover, the analysis points out which systems should be used depending on whether the user prefers to maximize the accuracy or the interpretability of the models.

UR - http://www.scopus.com/inward/record.url?scp=46949084846&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-78979-6_10

DO - 10.1007/978-3-540-78979-6_10

M3 - Chapter

AN - SCOPUS:46949084846

SN - 9783540789789

T3 - Studies in Computational Intelligence

SP - 205

EP - 230

BT - Learning Classifier Systems in Data Mining

A2 - Bull, Larry

A2 - Bernadó-Mansilla, Ester

A2 - Holmes, John

ER -

A comparative study of several genetic-based supervised learning systems

Abstract

Publication series

Access to Document

Other files and links

Fingerprint

Cite this