Fraud class histogram after oversampling.

<div><p>Credit card fraud is a significant problem that costs billions of dollars annually. Detecting fraudulent transactions is challenging due to the imbalance in class distribution, where the majority of transactions are legitimate. While pre-processing techniques such as oversampling...

Full description

Saved in:

Bibliographic Details
Main Author:	HaiChao Du (18114974) (author)
Other Authors:	Li Lv (382638) (author), Hongliang Wang (440441) (author), An Guo (1451287) (author)
Published:	2024
Subjects:	Genetics Cancer Science Policy Biological Sciences not elsewhere classified Information Systems not elsewhere classified true positive rate true negative rate matthew &# 8217 feature representation learning extract relevant features real dataset anonymized probabilistic xgboost based overgeneralized samples generated detecting fraudulent transactions acc index compared realistic data distributions novel method proposed often generate unrealistic method called autoencoder overgeneralized samples realistic data novel method promising method mcc index generate new also compared novel approach fraud data synthetic minority significant problem sampling technique processing techniques process effectively paper proposes overall improvement original dataset minority classes light gbm highly imbalanced higher accuracy dollars annually costs billions correlation coefficient classification according class distribution challenging due
Tags:	Add Tag No Tags, Be the first to tag this record!

Fraud class histogram after oversampling.

Similar Items