Cost-Sensitive and Leakage-Aware Evaluation of Credit Card Fraud Detection Under Extreme Class Imbalance

Abstract

Credit-card fraud detection is difficult because fraudulent transactions represent a very small

share of observed payments, while false negatives and false positives impose different

operational costs. This study presents a reproducible framework for evaluating fraud-detection

classifiers under extreme class imbalance, with explicit controls for data leakage, threshold

selection, and decision cost. The protocol covers dataset validation, partition design,

training-only preprocessing and resampling, baseline models, validation-based threshold

selection, and reporting of ranking, classification, calibration, and workload measures. It

prioritizes average precision, precision, recall, false-positive rate, confusion counts, and alert

volume rather than accuracy alone.

The public European Credit Card Fraud Detection benchmark is described as containing

284,807 transactions, including 492 fraud cases, for a fraud prevalence of approximately

0.1727%. From these published aggregate counts, an all-legitimate rule achieves

approximately 99.8273% accuracy while detecting no fraud. This analytical baseline

demonstrates why accuracy alone is inadequate. The transaction-level CSV was not available

in the execution environment used for this study; consequently, no trained-model comparison

or model-specific performance claim is presented. The contribution is a transparent evaluation

design and a verified calculation from published aggregate counts, not a ranking of fitted

classifiers. The paper also examines temporal validity, anonymized features, privacy, fairness,

calibration, and the boundary between benchmark results and production readiness.

Keywords: Credit-card fraud detection; extreme class imbalance; data leakage; cost-sensitive

evaluation; decision thresholds; precision–recall analysis.

Leakage-free evaluation of task-aware Nyström sampling for IoMT intrusion detection
Hiba A. Tarish, Rosilah Hassan, Khairul Akram Zainol Ariffin +1
Frontiers in artificial intelligence16 Oct 2026Peer-reviewedCC BYFull textEurope PMC
Leakage-Safe Machine Learning for Hydrogen Embrittlement Detection in 316L Stainless Steel: A Region-Held-Out Evaluation of Texture and Deep Features in SEM Micrographs
Muhammad Awais, Muhammad Yaseen, Abdul Shakoor +3
arXiv23 Sep 2026PreprintarXiv licenceFull textCondensed matterarXiv
Comparative evaluation of machine learning algorithms for Maize Lethal Necrosis Disease detection under class imbalance
Maurice Wanyonyi
PLOS One23 Oct 2026Peer-reviewedCC BYFull textPlant biologyPLOSEurope PMC
A Deep Learning Approach for Automated Re-Inspection of False Calls in Mixed-Production Surface Mount Lines Under Extreme Class Imbalance
Kun Tian, Zhanwen Niu
Applied Sciences26 Oct 2026Peer-reviewedCC BYFull textGeneral biology General chemistryDOAJ
Spaghettification: How extreme replichore imbalance impacts bacterial replication and morphology
S. Cao, F. Ilievski, D. L. Huseby +3
bioRxiv24 Sep 2026PreprintCC BY-NC-NDFull textMicrobiologybioRxiv

Cite

BibTeX RIS

BibTeX

@misc{khan2026costsensitive,
  title = {{Cost-Sensitive and Leakage-Aware Evaluation of Credit Card Fraud Detection Under Extreme Class Imbalance}},
  author = {Rayhan Khan and Reena Baghel and Reva Sharma and Roshni Singh},
  howpublished = {Zenodo},
  year = {2026},
  month = oct,
  doi = {10.5281/zenodo.23448825},
  url = {https://zenodo.org/records/23448825},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Cost-Sensitive and Leakage-Aware Evaluation of Credit Card Fraud Detection Under Extreme Class Imbalance
AU  - Rayhan Khan
AU  - Reena Baghel
AU  - Reva Sharma
AU  - Roshni Singh
PY  - 2026
DA  - 2026/10/18
PB  - Zenodo
DO  - 10.5281/zenodo.23448825
UR  - https://zenodo.org/records/23448825
AB  - Credit-card fraud detection is difficult because fraudulent transactions represent a very small  share of observed payments, while false negatives and false positives impose different  operational costs. This study presents a reproducible framework for evaluating fraud-detection  classifiers under extreme class imbalance, with explicit controls for data leakage, threshold  selection, and decision cost. The protocol covers dataset validation, partition design,  training-only preprocessing and resampling, baseline models, validation-based threshold  selection, and reporting of ranking, classification, calibration, and workload measures. It  prioritizes average precision, precision, recall, false-positive rate, confusion counts, and alert  volume rather than accuracy alone.  The public European Credit Card Fraud Detection benchmark is described as containing  284,807 transactions, including 492 fraud cases, for a fraud prevalence of approximately  0.1727%. From these published aggregate counts, an all-legitimate rule achieves  approximately 99.8273% accuracy while detecting no fraud. This analytical baseline  demonstrates why accuracy alone is inadequate. The transaction-level CSV was not available  in the execution environment used for this study; consequently, no trained-model comparison  or model-specific performance claim is presented. The contribution is a transparent evaluation  design and a verified calculation from published aggregate counts, not a ranking of fitted  classifiers. The paper also examines temporal validity, anonymized features, privacy, fairness,  calibration, and the boundary between benchmark results and production readiness.  Keywords: Credit-card fraud detection; extreme class imbalance; data leakage; cost-sensitive  evaluation; decision thresholds; precision–recall analysis.
ER  -

Abstract

Related

Cite