LibAUC: A Deep Learning Library for X-Risk Optimization
Detecting Frauds in Online Transactions using Anamoly Detection Techniques Such as Over Sampling and Under-Sampling as the ratio of Frauds is less than 0.00005 thus, simply applying Classification Algorithm may result in Overfitting
Trying to recogize and predict fraud in financial transactions is a good example of binary classification analysis. A transaction either is fraudulent, or it is genuine. What makes fraud detection especially challenging is the is the highly imbalanced distribution between positive (genuine) and negative (fraud) classes.
CheXpert-based chest X-ray multi-label classification research PoC with DenseNet121, uncertainty-label experiments, threshold tuning, and Grad-CAM explainability.
Model-agnostic geometric preprocessing for imbalanced tabular learning, with reproducible benchmarks, derived result artifacts, and paper-ready figures.
Cost-aware credit card fraud detection pipeline: time-based split, probability calibration, and business-aligned threshold tuning (AUPRC-first).
To associate your repository with the auprc topic, visit your repo's landing page and select "manage topics."
