GM4OS: An Evolutionary Oversampling Approach for Imbalanced Binary Classification Tasks

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Imbalanced datasets pose a significant and longstanding challenge to machine learning algorithms, particularly in binary classification tasks. Over the past few years, various solutions have emerged, with a substantial focus on the automated generation of synthetic observations for the minority class, a technique known as oversampling. Among the various oversampling approaches, the Synthetic Minority Oversampling Technique (SMOTE) has recently garnered considerable attention as a highly promising method. SMOTE achieves this by generating new observations through the creation of points along the line segment connecting two existing minority class observations. Nevertheless, the performance of SMOTE frequently hinges upon the specific selection of these observation pairs for resampling. This research introduces the Genetic Methods for OverSampling (GM4OS), a novel oversampling technique that addresses this challenge. In GM4OS, individuals are represented as pairs of objects. The first object assumes the form of a GP-like function, operating on vectors, while the second object adopts a GA-like genome structure containing pairs of minority class observations. By co-evolving these two elements, GM4OS conducts a simultaneous search for the most suitable resampling pair and the most effective oversampling function. Experimental results, obtained on ten imbalanced binary classification problems, demonstrate that GM4OS consistently outperforms or yields results that are at least comparable to those achieved through linear regression and linear regression when combined with SMOTE.
Original languageEnglish
Title of host publicationApplications of Evolutionary Computation
Subtitle of host publication27th European Conference, EvoApplications 2024, Held as Part of EvoStar 2024, Aberystwyth, UK, April 3–5, 2024, Proceedings, Part I
EditorsStephen Smith, João Correia, Christian Cintrano
Place of PublicationCham, Switzerland
PublisherSpringer Nature Switzerland AG
Pages68-82
Number of pages15
Volume1
ISBN (Electronic)978-3-031-56852-7
ISBN (Print)978-3-031-56851-0
DOIs
Publication statusPublished - 21 Apr 2024
Event27th International Conference on the Applications of Evolutionary Computation, held as part of EvoStar 2024 - Aberystwyth University, Aberystwyth, United Kingdom
Duration: 3 Apr 20245 Apr 2024
Conference number: 27
https://www.evostar.org/2024/evoapps/

Publication series

NameLecture Notes in Computer Science
PublisherSpringer Nature Switzerland AG
Volume14634
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference27th International Conference on the Applications of Evolutionary Computation, held as part of EvoStar 2024
Abbreviated titleEvoApplications
Country/TerritoryUnited Kingdom
CityAberystwyth
Period3/04/245/04/24
Internet address

Keywords

  • Oversampling
  • Imbalanced Data
  • Binary Classification
  • Genetic Programming
  • Genetic Algorithms

Fingerprint

Dive into the research topics of 'GM4OS: An Evolutionary Oversampling Approach for Imbalanced Binary Classification Tasks'. Together they form a unique fingerprint.

Cite this