Implementing Machine Learning Algorithms To Enhance Intrusion Detection System Across Computerised Networks Towards Pre-empting Cyber Attacks

Garg, Shruti Praveen (2024) Implementing Machine Learning Algorithms To Enhance Intrusion Detection System Across Computerised Networks Towards Pre-empting Cyber Attacks. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (846kB) | Preview
Preview	PDF (Configuration Manual) Download (804kB) | Preview

Abstract

The development of cyber attacks has significant threat on the security of any networked system. This research highlights the advancement of machine learning to enhance Intrustion Detection System for detection of unforeseen cyber attacks, detection and mitigation of a variety of network intrusions namely, Distributed Denial of Service Attack, Brute Force, SQL Injection, etc. on corporate network. With the use of CIC-IDS 2017 dataset which resembles real-world data, the study compares the execution of multiple models like Logistic Regression, Random Forest, Isolation Forest, and OneClass SVM individually and combines them into a Voting Classifier using soft voting with minimizing false positives and detecting accuracy being the pivotal part of the scale. Following a sequence of preprocessing steps which included feature scaling, label encoding, and missing data imputation, the dataset was assembled for model training. To understand the relevant characteristics for intrusion detection, feature selection techniques were applied, and the models were trained and assessed using classification metrics, precision, recall, F1-score, and ROC-AUC for an extensive assessment of the model performances. Ensemble mechanism- Voting Classifier was implemented in combining the prediction of all the models to magnify the detection accuracy. The results display that a combination of ensemble-based technique with supervised learning and anomaly detection technique gives more advanced performance in detecting network anomalies when compared with individual models. This study highlights the ability of machine learning approaches in enhancing the reliability and effectiveness of intrusion detection system also offers insights to secure corporate network from advanced cyberattacks.

Item Type:	Thesis (Masters)
Supervisors:	Name Email McCabe, Liam UNSPECIFIED
Uncontrolled Keywords:	Intrusion detection system; machine learning; data preprocessing; model; training; classification metrices; precision; recall; F1-score; Logistic Regression; Deep Neural Network; Isolation Forest; One-Class SVM; Random Forest; Voting Classifier
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 Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	18 Jul 2025 11:37
Last Modified:	18 Jul 2025 11:37
URI:	https://norma.ncirl.ie/id/eprint/8205

Actions (login required)

View Item