Toshi Bhat

Publication Date

Spring 2024

Degree Type

Master's Project

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

First Advisor

William Andreopoulos

Second Advisor

Robert Chun

Third Advisor

Chung-Wen Tsao

Keywords

Credit Card Fraud Detection, AI, Enhanced User Interaction

Abstract

This thesis describes the development and testing of a unique system for detecting credit card fraud. The system employs graph neural networks (GNNs) and a real-time user interaction platform. The primary goal of this study is to use advanced machine learning methods and interactive technologies to improve fraud detection accuracy and the speed with which users can receive assistance. GraphSAGE, a type of GNN, was trained on a simulated set of credit card transactions, allowing the system to detect and predict fraud very accurately. Simulating a real-world transaction scenario is an important aspect of the project. In this case, the system processes real-time data and communicates with users via a chatbot. This bot was created using the Hugging Face Transformers library. Its job is to communicate directly with users and use their feedback to improve the model’s ability to predict potentially fraudulent transactions. Experiments show that the GraphSAGE model outperforms traditional machine learning models in terms of detecting fraudulent transactions. It has higher levels of accuracy and recall. This study advances the field of fraud detection by demonstrating how a synergized system with GNNs work with systems that receive feedback from users in real time. It teaches us useful information about how these technologies can be used in banking and provides ideas for combating credit card fraud that can be expanded as the problem worsens. In the future, this method may be used to combat other types of financial fraud. And as interactivity and machine learning improve, it may work even better.

Download

Available for download on Friday, May 23, 2025

Share

COinS