Weighted Cross-Entropy to tackle Overlapping in Fraud Detection - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Weighted Cross-Entropy to tackle Overlapping in Fraud Detection

Claire Verdier
  • Fonction : Auteur
  • PersonId : 1175194
Stephane Perriot
  • Fonction : Auteur
Arnault Pachot
  • Fonction : Auteur
  • PersonId : 1175890

Résumé

This paper proposes a new loss function for unbalanced binary data sets with overlapping classes. This loss function is a weighted cross-entropy binary function that takes as its argument the distance between an observation of the majority class and an observation of the nearest minority class. The distance is calculated using the nearest neighbor algorithm. The quality of this loss function is measured with the AUC-ROC metric on a fraud detection data set for energy saving certificates. In addition, the reduction in the workload for human verification is calculated. The new loss function improves the result. A statistical test confirms the improvement of the model using the new loss function.
Fichier principal
Vignette du fichier
ArticleFraudeDetection.pdf (354.87 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03832865 , version 1 (28-10-2022)

Identifiants

  • HAL Id : hal-03832865 , version 1

Citer

Claire Verdier, Stephane Perriot, Arnault Pachot. Weighted Cross-Entropy to tackle Overlapping in Fraud Detection. 15th International Conference on Machine Learning and Computing, Feb 2023, Zhuhai, China. ⟨hal-03832865⟩
137 Consultations
48 Téléchargements

Partager

Gmail Facebook X LinkedIn More