Fraud Detection Project

Overview

This project aims to detect fraudulent transactions using a dataset sourced from Kaggle. Multiple machine learning classifiers are employed to predict fraudulent activity. To address the class imbalance issue inherent in fraud detection, SMOTE (Synthetic Minority Over-sampling Technique) is incorporated to augment the training data and improve model performance.

Features

• Input a list of classifiers to evaluate their performance.
• SMOTE integration to handle class imbalance.
• Model training and evaluation on imbalanced datasets.
• Easy to extend with additional classifiers or techniques.

Tools Used

• Python
• Scikit-learn
• Imbalanced-learn (SMOTE)
• Pandas
• NumPy
• Matplotlib (for visualizations)
• XGBoost

More.................... [In Progress]