Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting

Danang ; Danang ; Toni Wijanarko Adi Putra
International Journal of Information Technology an... 📅 2025 Vol 8, No 1 🔗 DOI: 10.24246/ijiteb.812025.13-24
Abstract
Fraud detection on payment transactions is an extremely imbalanced, high-stakes classification task in which deployment decisions depend not only on ranking quality but also on reliable probability estimates. We study credit card fraud detection on a standard real-transaction benchmark (284,807 transactions; 492 frauds) and target two deployment requirements: cost-sensitive thresholding under asymmetric error costs and reliability calibration so model outputs can be interpreted as stable risk scores. We benchmark logistic regression and XGBoost and propose a focal-proxy reweighting scheme for boosted trees via iterative weight updates inspired by focal loss. Probabilities are calibrated on validation using Platt scaling, temperature scaling, and isotonic-style monotone calibration; the best calibrator is selected by minimum validation Brier score. For decision-making, we choose the operating threshold that minimizes expected cost, Cost(t) = 10 · FN(t) + 1 · FP(t), on validation, then evaluate on a held-out test set. On the benchmark split (train 199,364; validation 42,721; test 42,722), the calibrated XGBoost baseline achieves AUROC 0.973, AUPRC 0.812, fraud-class F1 0.767, and expected cost 154 with very low calibration error (ECE = 1.1 × 10⁻⁴). Overall, calibration reduces ECE and improves or maintains the Brier score, while cost-aware thresholding makes the FN/FP trade-off explicit via decision curves. 
Keywords
Class Imbalance; Cost-Sensitive Learning; Fraud Detection; Probability Calibration; Xgboost
How to Cite

Danang, et al. (2025). Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting. International Journal of Information Technology and Business (IJITEB), 8(1). https://doi.org/10.24246/ijiteb.812025.13-24

Danang, Danang; Toni Wijanarko Adi Putra, "Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting," International Journal of Information Technology and Business (IJITEB), vol. 8, no. 1, 2025.

Danang, Danang; Toni Wijanarko Adi Putra. "Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting." International Journal of Information Technology and Business (IJITEB), vol. 8, no. 1, 2025.

Danang, Danang; Toni Wijanarko Adi Putra. "Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting." International Journal of Information Technology and Business (IJITEB) 8, no. 1 (2025).

Danang, et al. (2025) 'Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting', International Journal of Information Technology and Business (IJITEB), 8(1). doi: 10.24246/ijiteb.812025.13-24.

Danang, Danang; Toni Wijanarko Adi Putra. Cost-Sensitive Fraud Detection with Reliability Calibration: A Practical Pipeline with XGBoost and Focal-Proxy Reweighting. International Journal of Information Technology and Business (IJITEB). 2025;8(1).

Artikel Terkait

Resilience of Deep Q-Network (DQN) Agent in Mitigating Ethereum Trading Risks Under Bearish Market Conditions

Adha, Fajri; Widiastuti, Tiwuk; Selvian Djahi, Bertha

An Adaptive AI-Driven Copywriting Framework: Design, Implementation, and Evaluation of a Web-Based GPT-Integrated Content Generation System

Daru, April Firman; Christanto, Febrian Wahyu; Prathivi, Rastri; Prasetyo, Dimas; Firdaus, Eryan Ahmad

Multi-Objective Reinforcement Learning for Supply Chain Optimization in Indonesia’s Free Nutritious Meal Program

Yanto, Budi; Saragih, Rusmin; Lubis, Adyanata; Elyandri Prasiwiningrum; Wahyuny, Romy

Modeling Factors Affecting Vocational School Students’ Perceived Ease of Use of Educational Technology

Hafid, Masruri; Krismiyati, Krismiyati; Diana, Andrushia

Determination of Maintenance Priority Based on Analytical Hierarchy Process for Magnetic Resonance Imaging

Nurdono Nurdono; Muhamad Haddin

Customer Loyalty Analysis Using RFM Model and K-Means Clustering for Marketing Strategy Optimization

Sahertian, Vigo Yano; Yessica Nataliani
Tren Sitasi Jurnal