AI-Based Fraud Detection Using Graph Neural Networks

Abstract

Fraud detection is an important problem in today’s digital world due to the rapid increase in online transactions. Traditional methods often fail to identify complex patterns of fraud. This project proposes an AI-based fraud-detection system using Graph Neural Networks (GNNs) to improve accuracy. In this system, transaction data is represented as a graph where nodes represent users or accounts and edges represent transactions between them, and a GraphSAGE-based GNN analyses these relationships to detect unusual patterns and identify fraudulent activities effectively. The system is implemented in Python using PyTorch and PyTorch Geometric for the GNN, with NetworkX for graph construction, scikit-learn for preprocessing, and a Streamlit interface for visualisation and prediction. It models a user–merchant transaction graph for edge-level fraud classification, incorporates realistic fraud indicators such as unusual transaction timing, device changes, and abnormal behaviour, supports uploading real UPI transaction CSVs or using built-in synthetic data, and displays risk scores together with explanations. The approach is designed to handle large amounts of data and provide faster, more reliable fraud detection, reducing false alerts while strengthening decision-making in real-time financial systems. The prototype was validated through functional test cases for dataset upload, missing-value handling, model training, fraud prediction, and visualisation, all of which passed. Overall, the system demonstrates how graph-based deep learning can significantly enhance fraud detection, making it more robust, scalable, and suitable for real-world financial applications.