Credit Card Fraud Detection

The increasing frequency of credit card fraud has led to significant financial losses globally.This project aims to develop a machine learning-based system for detecting fraudulent transactions using Logistic Regression.The system processes transaction data through preprocessing techniques like feature engineering, normalisation, and categorical encoding.To address the class imbalance issue, oversampling methods are applied.The model is trained to classify transactions as either legitimate or fraudulent with a focus on minimising false positives and false negatives.Model performance is evaluated using metrics such as accuracy, precision, recall, and F1-score.The system is designed for scalability and real-time detection, offering a robust solution to enhance financial security.It also lays the foundation for future upgrades involving deep learning or adaptive fraud detection techniques.