Evidential Undersampling Approach for Imbalanced Datasets with Class-Overlapping and Noise - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Evidential Undersampling Approach for Imbalanced Datasets with Class-Overlapping and Noise

Résumé

The class imbalance issue involves many real world domains such as fraud detection, medical diagnosis, intrusion detection, etc. Most classification algorithms tend to perform poorly when the training dataset is class-imbalanced. This problem gets more challenging in the presence of other factors such class-overlapping and noise. Among many methods, undersampling is a simple and efficient approach which rebalances the imbalanced dataset by removing majority samples. In this paper, we propose a novel method named Evidential Undersampling (EVUS), which is a re-sampling approach based on the theory of evidence. To avoid removing meaningful samples, each majority object is assigned a soft evidential label to gain more information about its location, then majority samples which are considered ambiguous or noisy by our framework, are eliminated from the training set. The conducted results with CART and SVM show that our proposal outperformed other well-known undersampling methods according to the AUC metric.
Fichier principal
Vignette du fichier
mdai2021_faresgr_vf.pdf (324.83 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03764122 , version 1 (30-08-2022)

Identifiants

Citer

Fares Grina, Zied Elouedi, Eric Lefevre. Evidential Undersampling Approach for Imbalanced Datasets with Class-Overlapping and Noise. International Conference on Modeling Decisions for Artificial Intelligence, MDAI'2021, Sep 2021, Umeå, Sweden. pp.181-192, ⟨10.1007/978-3-030-85529-1_15⟩. ⟨hal-03764122⟩

Collections

UNIV-ARTOIS LGI2A
12 Consultations
88 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More