NORMA eResearch @NCI Library

In-Depth Analysis of Machine Learning for Securing Internet of Things devices using CIC IoT & Net Flow Dataset

Srinivasa, Shreyas (2024) In-Depth Analysis of Machine Learning for Securing Internet of Things devices using CIC IoT & Net Flow Dataset. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (441kB) | Preview
[thumbnail of Configuration Manual]
Preview
PDF (Configuration Manual)
Download (367kB) | Preview

Abstract

The rise of Internet of Things (IoT) has made the digital landscape transformed into providing more services but also introduced significant cybersecurity challenges by expanding potential vulnerabilities. Security systems such as Data Loss Prevention (DLP), Intrusion Detection Systems (IDS) and firewalls are struggling to keep up with these modern threats. They often produce a high number of false positives and lack the capability to identify more advanced, evolving attacks. To address these shortcomings, our research presents a machine learning models where we compare the ML models and analyse in detail as to which model produces the best accuracy. This approach has been documented each model and made sure the results are best for the networks for processing sequential data. The hybrid design improves detection accuracy and reduces the rate of false positives.

Furthermore, we developed these machine learning models are two datasets in order to figure out the best results. The novelty of this research lies in the hyper parameter tuning of machine learning models to achieve best results. The main contributions of this research are the extensive machine learning and deep learning algorithms for detecting and classifying malicious traffic. Our research conducted five novel machine learning models on UQ-NIDS dataset and added hyper parameter tuning in order to determine how well the model performs on those IoT attacks and for CICIoT dataset, our research is performed extensively on the classification of attack categories and implemented three novel algorithms to find the best model for each attack categories. And our research was able to find that Random Forest and Logistic Regression performs well on both the datasets and also identifying different categories in IoT attacks.

Item Type: Thesis (Masters)
Supervisors:
Name
Email
Moldovan, Arghir Nicolae
UNSPECIFIED
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
Q Science > QA Mathematics > Computer software > Computer Security
T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > Computer networks > Internet of things
Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions: School of Computing > Master of Science in Cyber Security
Depositing User: Ciara O'Brien
Date Deposited: 28 Jul 2025 11:25
Last Modified: 28 Jul 2025 11:25
URI: https://norma.ncirl.ie/id/eprint/8262

Actions (login required)

View Item View Item