The class imbalance problem in UCS classifier system: A preliminary study

Albert Orriols-Puig; Ester Bernadó-Mansilla

doi:10.1007/978-3-540-71231-2_12

The class imbalance problem in UCS classifier system: A preliminary study

Albert Orriols-Puig^*
, Ester Bernadó-Mansilla

^*Corresponding author for this work

Research output: Book chapter › Conference contribution › peer-review

7 Citations (Scopus)

Abstract

The class imbalance problem has been said recently to hinder the performance of learning systems. In fact, many of them are designed with the assumption of well-balanced datasets. But this commitment is not always true, since it is very common to find higher presence of one of the classes in real classification problems. The aim of this paper is to make a preliminary analysis on the effect of the class imbalance problem in learning classifier systems. Particularly we focus our study on UCS, a supervised version of XCS classifier system. We analyze UCS's behavior on unbalanced dataseis and find that UCS is sensitive to high levels of class imbalance. We study strategies for dealing with class imbalances, acting either at the sampling level or at the classifier system's level.

Original language	English
Title of host publication	Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers
Publisher	Springer Verlag
Pages	161-180
Number of pages	20
ISBN (Print)	9783540712305
DOIs	https://doi.org/10.1007/978-3-540-71231-2_12
Publication status	Published - 2007

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	4399 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Access to Document

10.1007/978-3-540-71231-2_12

Cite this

Orriols-Puig, A., & Bernadó-Mansilla, E. (2007). The class imbalance problem in UCS classifier system: A preliminary study. In Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers (pp. 161-180). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4399 LNAI). Springer Verlag. https://doi.org/10.1007/978-3-540-71231-2_12

Orriols-Puig, Albert ; Bernadó-Mansilla, Ester. / The class imbalance problem in UCS classifier system : A preliminary study. Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers. Springer Verlag, 2007. pp. 161-180 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{3ce36bb95907403a8a88053f36489f0a,

title = "The class imbalance problem in UCS classifier system: A preliminary study",

abstract = "The class imbalance problem has been said recently to hinder the performance of learning systems. In fact, many of them are designed with the assumption of well-balanced datasets. But this commitment is not always true, since it is very common to find higher presence of one of the classes in real classification problems. The aim of this paper is to make a preliminary analysis on the effect of the class imbalance problem in learning classifier systems. Particularly we focus our study on UCS, a supervised version of XCS classifier system. We analyze UCS's behavior on unbalanced dataseis and find that UCS is sensitive to high levels of class imbalance. We study strategies for dealing with class imbalances, acting either at the sampling level or at the classifier system's level.",

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

year = "2007",

doi = "10.1007/978-3-540-71231-2\_12",

language = "English",

isbn = "9783540712305",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "161--180",

booktitle = "Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers",

address = "Germany",

}

Orriols-Puig, A & Bernadó-Mansilla, E 2007, The class imbalance problem in UCS classifier system: A preliminary study. in Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4399 LNAI, Springer Verlag, pp. 161-180. https://doi.org/10.1007/978-3-540-71231-2_12

The class imbalance problem in UCS classifier system: A preliminary study. / Orriols-Puig, Albert; Bernadó-Mansilla, Ester.
Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers. Springer Verlag, 2007. p. 161-180 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4399 LNAI).

Research output: Book chapter › Conference contribution › peer-review

TY - GEN

T1 - The class imbalance problem in UCS classifier system

T2 - A preliminary study

AU - Orriols-Puig, Albert

AU - Bernadó-Mansilla, Ester

PY - 2007

Y1 - 2007

N2 - The class imbalance problem has been said recently to hinder the performance of learning systems. In fact, many of them are designed with the assumption of well-balanced datasets. But this commitment is not always true, since it is very common to find higher presence of one of the classes in real classification problems. The aim of this paper is to make a preliminary analysis on the effect of the class imbalance problem in learning classifier systems. Particularly we focus our study on UCS, a supervised version of XCS classifier system. We analyze UCS's behavior on unbalanced dataseis and find that UCS is sensitive to high levels of class imbalance. We study strategies for dealing with class imbalances, acting either at the sampling level or at the classifier system's level.

AB - The class imbalance problem has been said recently to hinder the performance of learning systems. In fact, many of them are designed with the assumption of well-balanced datasets. But this commitment is not always true, since it is very common to find higher presence of one of the classes in real classification problems. The aim of this paper is to make a preliminary analysis on the effect of the class imbalance problem in learning classifier systems. Particularly we focus our study on UCS, a supervised version of XCS classifier system. We analyze UCS's behavior on unbalanced dataseis and find that UCS is sensitive to high levels of class imbalance. We study strategies for dealing with class imbalances, acting either at the sampling level or at the classifier system's level.

UR - https://www.scopus.com/pages/publications/38049103501

U2 - 10.1007/978-3-540-71231-2_12

DO - 10.1007/978-3-540-71231-2_12

M3 - Conference contribution

AN - SCOPUS:38049103501

SN - 9783540712305

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 161

EP - 180

BT - Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers

PB - Springer Verlag

ER -

Orriols-Puig A, Bernadó-Mansilla E. The class imbalance problem in UCS classifier system: A preliminary study. In Learning Classifier Systems - International Workshops, IWLCS 2003-2005, Revised Selected Papers. Springer Verlag. 2007. p. 161-180. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-540-71231-2_12

The class imbalance problem in UCS classifier system: A preliminary study

Abstract

Publication series

Access to Document

Other files and links

Fingerprint

Cite this