Exploring data augmentation and active learning benefits in imbalanced datasets

Vista sencilla de ítem
dc.contributor.authorMoles, Luis
dc.contributor.authorAndres, Alain
dc.contributor.authorEchegaray, Goretti
dc.contributor.authorBoto Sánchez, Fernando
dc.date.accessioned2024-11-12T13:15:53Z
dc.date.available2024-11-12T13:15:53Z
dc.date.issued2024-06
dc.date.updated2024-11-12T13:15:53Z
dc.description.abstractDespite the increasing availability of vast amounts of data, the challenge of acquiring labeled data persists. This issue is particularly serious in supervised learning scenarios, where labeled data are essential for model training. In addition, the rapid growth in data required by cutting-edge technologies such as deep learning makes the task of labeling large datasets impractical. Active learning methods offer a powerful solution by iteratively selecting the most informative unlabeled instances, thereby reducing the amount of labeled data required. However, active learning faces some limitations with imbalanced datasets, where majority class over-representation can bias sample selection. To address this, combining active learning with data augmentation techniques emerges as a promising strategy. Nonetheless, the best way to combine these techniques is not yet clear. Our research addresses this question by analyzing the effectiveness of combining both active learning and data augmentation techniques under different scenarios. Moreover, we focus on improving the generalization capabilities for minority classes, which tend to be overshadowed by the improvement seen in majority classes. For this purpose, we generate synthetic data using multiple data augmentation methods and evaluate the results considering two active learning strategies across three imbalanced datasets. Our study shows that data augmentation enhances prediction accuracy for minority classes, with approaches based on CTGANs obtaining improvements of nearly 50% in some cases. Moreover, we show that combining data augmentation techniques with active learning can reduce the amount of real data required.en
dc.description.sponsorshipThis work was financed by the Basque Government through their Elkartek program (SONETO project, ref. KK-2023/00038)en
dc.identifier.citationMoles, L., Andres, A., Echegaray, G., & Boto, F. (2024). Exploring Data Augmentation and Active Learning Benefits in Imbalanced Datasets. Mathematics, 12(12). https://doi.org/10.3390/MATH12121898
dc.identifier.doi10.3390/MATH12121898
dc.identifier.issn2227-7390
dc.identifier.urihttp://hdl.handle.net/20.500.14454/1782
dc.language.isoeng
dc.publisherMultidisciplinary Digital Publishing Institute (MDPI)
dc.rights© 2024 by the authors
dc.subject.otherActive learning
dc.subject.otherCTGAN
dc.subject.otherData augmentation
dc.subject.otherEntropy sampling
dc.subject.otherMachine learning
dc.titleExploring data augmentation and active learning benefits in imbalanced datasetsen
dc.typejournal article
dcterms.accessRightsopen access
oaire.citation.issue12
oaire.citation.titleMathematics
oaire.citation.volume12
oaire.licenseConditionhttps://creativecommons.org/licenses/by/4.0/
oaire.versionVoR
Ficheros en el ítem
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
moles_exploring_2024.pdf
Tamaño:
12.39 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Artículos