Intrusion Detection using Machine Learning with Real-Time Dashboard

Mora, Sangeetha (2024) Intrusion Detection using Machine Learning with Real-Time Dashboard. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (814kB) | Preview
Preview	PDF (Configuration Manual) Download (784kB) | Preview

Abstract

Every day that passes sees new, more complex, and developed cyber threats thus requiring strong IDS that offers real-time detection and analysis. This study aims to design an IDS based on machine learning to identify abnormal occurrences and directly display the result, which can effectively increase cybersecurity defence and improve operations.

The study employs the UNSW-NB15 dataset and first and second-level system and firewall data of an Ubuntu operating system-based virtual machine. In classification tasks, Decision Trees, Random Forests, and Gaussian Naive Bayes are used to classify the feature vectors appropriately. For the process of feature selection, Recursive Feature Elimination is applied whereas for anomaly detection in the user’s authentication data Isolation Forest Algorithm is used. Logistic Regression is being used on firewall data for penetration prediction. Data preprocessing takes care of the appropriate arrangement of the gathered data and applying qualitative analysis.

According to the model, the highest accuracy of classification was 95.01% with the Random Forest model enabling measurement of high precision and recall. The analysis of the firewall logs using Logistic Regression yielded an accuracy of around 88.68%, with the F1 score touching the balance to ensure a better focus on the network intrusions. Some of the insights are fed into another real-time threat that will be constantly monitoring threats.

The proposed method fills the gap between formal and intuitive analysis of logs with current systems of machine learning in which the approach is not only scalable and efficient but adaptable to real-time intrusion objectives. Further improvements will target at including advanced technologies (e.g., deep learning), adding real-time alert sender, and these kinds of techniques to the distributed system and so forth.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Cosgrave, Noel UNSPECIFIED
Uncontrolled Keywords:	Decision Trees; Random Forests; Gaussian Naive Bayes; Recursive Feature Elimination; firewall data; Authentication data; Intrusion Detection System
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:	03 Sep 2025 14:23
Last Modified:	03 Sep 2025 14:23
URI:	https://norma.ncirl.ie/id/eprint/8752

Actions (login required)

View Item