RCSMOTE: Range-Controlled Synthetic Minority Over-sampling Technique for handling the class imbalance problem

Detecting credit card fraud using a neural network model on highly imbalanced classes.

A robust end-to-end machine learning pipeline for credit card fraud detection using Python, scikit-learn, and Streamlit. Includes data preprocessing, feature selection, model training & evaluation, saving the best model, and an interactive Streamlit app for predictions.

Image classification pipeline relying on Deep Learning models dealing with training over multiple datasets with unbalanced labels and classes. In addition, also classification using histogram data was done for better generalization over different datasets.

Predicting strokes using neural networks with a focus on imbalanced datasets.

My take on "Do not overfit! II" competition on Kaggle which challenges participants to avoid overfitting.

This project employs advanced SMOTE variants to effectively work on imbalanced classification challenges in machine learning datasets.

Credit Card Ownership Prediction A machine learning project that predicts credit card ownership using features like age and income, balancing class distributions for improved accuracy.