Path and Cost Based GNN Model for Social Media Network Fraud Detection

Abstract

Detecting fraud in social media networks is a growing challenge due to the evolving nature of fraudulent activities. Traditional methods often fail to capture the intricate relationships among users, making fraud detection less effective. This study introduces a Graph Neural Network (GNN)-based model to identify fraudulent activities by leveraging the structural properties of social media interactions. In this approach, users and their relationships are represented as graph nodes and edges, allowing the model to learn hidden patterns and dependencies. By utilizing GNNs, the system effectively distinguishes between genuine and fraudulent behaviours. The model is trained and tested on real-world datasets, demonstrating superior performance in terms of accuracy, precision, recall, and F1-score compared to conventional machine learning techniques. The findings suggest that GNNs offer a powerful and scalable solution for fraud detection in social media networks, enhancing security and reliability