Theses and Dissertations

Date of Award

8-2025

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Applied Statistics and Data Science

First Advisor

Hansapani Rodrigo

Second Advisor

Sanjeev Kumar

Third Advisor

George P. Yanev

Abstract

Cybersecurity is known today as one of the greatest challenges of the modern era. Among the various types of cyber-attacks that threaten our security, the Distributed Denial of Service (DDoS) attack is among some of the most common, effective, and well-recognized attack strategies. Since this form of attack is meant to disrupt the availability factor covertly, it can be detrimental to the targeted machines and difficult to discover. Because of that, there have been several approaches, as well as solutions that have been devised to detect it as accurately and efficiently as possible. In this study, four sequential data modeling techniques: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRUs), Transformers and two (2) non-sequential data modeling techniques: Random Forest (RF) and Deep Neural Network (DNN) have been used to build the intrusion detection system. The CIC-DDoS-2019 dataset was utilized in this work for training and testing the performance of the models. In this work, we concentrated on addressing data imbalance issues, which arise from the high volume of attack data compared to benign data. The problem of data imbalance was addressed using five distinct data balancing techniques: four oversampling and one undersampling, after which the performance of both sequential and nonsequential models was evaluated. The performance of these models was measured using precision, recall, F1 score, balanced accuracy, and AUC score, across each of the data balancing strategies implemented.

Comments

Copyright 2025 Vincent Agbenyeavu. All Rights Reserved. https://proquest.com/docview/3253957302

Download

Share

COinS