Variational Autoencoder(VAE) for Anomaly Detection in Network traffic

Sayyad, Sharik Arif (2024) Variational Autoencoder(VAE) for Anomaly Detection in Network traffic. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (829kB) | Preview
Preview	PDF (Configuration Manual) Download (493kB) | Preview

Abstract

This is vital as well, to find out the early vulnerabilities and threats in cyberspace eco-system respectively, known we commonly call IDS that stands for Intrusion Detection Systems. However, legacy intrusion detection systems (IDS) frameworks are often ill-equipped to handle the dynamic and complex nature of today's advanced cyberattacks which will need better solutions. In this work, we investigate the application of autoencoder models to enhance intrusion detection systems (IDS), as they are known for their performance in anomaly detection. We designed and evaluated five different autoencoder architectures: a basic one, a convolutional (ConvAE), Variational AutoEncoder(VAE), Conditional VAE and an Adversarial AE. This approach was deployed to balance the class imbalance in two complex network datasets that were used both as training and testing sets for each model by means of SMOTE method. The results showed that the Convolutional Variational Autoencoder (CVAE) outperformed other models with almost perfect scores in accuracy, precision and recall among all models as shown by F1- scores. This places the CVAE in high regard as a network traffic classifier, given its superiority over prior methods for solidifying benign and malicious networking distinction. The study results suggests that combining deep learning CVAE architectures in Intrusion Detection Systems (IDS) can result on strong networks protection against many computer network attacks as well.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Menghwar, Teerath Kumar UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2025 11:03
Last Modified:	25 Aug 2025 11:03
URI:	https://norma.ncirl.ie/id/eprint/8624

Actions (login required)

View Item