votrananhquan/fraud-detection-model

Credit Card Fraud Detection — LightGBM Champion Model

Model Description

LightGBM (lgbm_large) champion model for credit card fraud detection on the Kaggle Credit Card Fraud Dataset. Registered in MLflow Registry as fraud-detection-model@production.

Performance (Test Set — 56,962 samples)

Metric	Value
PR-AUC	0.8770
Recall	0.8571
Precision	0.8485
F1	0.8528

Usage

import joblib
import numpy as np

model = joblib.load("fraud_model.pkl")
# features: V1-V28 (PCA), Amount (StandardScaler μ=88.35, σ=250.12)
X = np.array([[...]])  # shape (1, 29)
proba = model.predict_proba(X)[:, 1]  # fraud probability

MLflow Tracking

• Run ID: 5b382ed3ae21492788e8494404474852
• Model version: 2
• Registered name: fraud-detection-model