Enhancing Network Security Using Machine Learning Model-Agnostic Approach on Diverse Datasets

Zaeem, Muhammad (2024) Enhancing Network Security Using Machine Learning Model-Agnostic Approach on Diverse Datasets. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (677kB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

With the evolution of technology, everything has moved to online channels making it easier for hackers and other cyber threats to make intrusion attacks and cause the breach of personal or official data. This has increased the demand for more secure and updated intrusion detection systems (IDS). Firewalls serve as the first line of defence against cyber-attacks however, their signature-based architecture fails to defend against newer attacks. This study proposes using machine learning algorithms with current firewalls to increase network classification and intrusion detection efficiency. The research incorporates three different network classification and intrusion detection datasets that are analysed using a model-agnostic pipeline comprising of logistic regression, naïve bayes, random forest, XGBoost, KNN, SVM and ANN while exploring challenges like outlier analysis, data imbalance and feature diversity. The Grid Search algorithm is used for the efficient hyperparameter tuning of all the models. All models' results are comparatively evaluated to find the best machine learning model to deal with the intrusion problem. The performance of all models was found to increase after oversampling of minority classes, feature selection and hyperparameter tuning of models. Random Forest and XGBoost emerged as the best-performing models with an F1-score of 0.99 on two datasets and 0.80 on the third dataset.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Chikkankod, Arjun 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 Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	26 Aug 2025 12:30
Last Modified:	26 Aug 2025 12:30
URI:	https://norma.ncirl.ie/id/eprint/8651

Actions (login required)

View Item