ANALYZING AND SAFEGUARDING FINANCIAL TRANSACTIONS AGAINST FRAUD

Abstract

This project focuses on developing a machine learning-based fraud detection system using the Paysim dataset to
identify fraudulent online payment transactions. Traditional fraud detection methods rely on static rules, which
fail to adapt to evolving fraud tactics. To address this, we propose a dynamic, data-driven approach using the
Decision Tree Classifier. The systemclassifies transactions as fraudulent or legitimate based on historical data,
leveraging features such as transaction amount, location, and user behavior.
Data preprocessing techniques like handling missing values, feature scaling, and encoding categorical data
were applied to prepare the dataset for training. The DecisionTreeClassifier model was trained and evaluated
using metrics such as accuracy, precision, recall, and F1- score. The model achieved high accuracy and was
capable of detecting fraud with minimal false positives. Additionally, the system integrates real-time fraud alerts
and performance metrics, allowing administrators to monitor transactions and model performance.
Future improvements could include integrating advanced machine learning models, such as Random Forests
and Gradient Boosting, to enhance accuracy. Moreover, handling real-time data and continuous model updates
will ensure the system remains effective against emerging fraud tactics. This project demonstrates the potential of
machine learning in creating robust and scalable fraud detection systems.