Tools
Tools
Porkodian Suganraj, Deepika (2023) Information Security and Data protection: A review of challenges and influencing factor faced in IT. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (2MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (2MB) | Preview |
Abstract
Information security and data protection are critical concerns in IT Systems, due to a rising risk of data breaches, abnormal network activity, human error, and privacy violations. The aim of this study is to examine the method and approach that may be employed to successfully lower the risks related to data breaches, anomalous network activity, and privacy violations in IT systems. A survey was also carried out among IT professionals to effectively find their understanding on data handling, incident reporting, information security, and other aspects on security, resulting in Intrusion detection systems (IDS) as the proposed approach to reduce the risks of privacy violations and data leaks in IT systems. The Deep Belief Network-Gated Recurrent Unit (DBN-GRU) architecture-based novel intrusion detection system makes use of a machine learning algorithm and the NSL-KDD99 dataset to identify intrusions. The method includes extensive data preprocessing to improve the quality of the data, such as binary encoding, data scaling, and Min-Max normalisation. Combining the sequential pattern recognition of GRUs with the hierarchical characteristic learning of DBNs, the hybrid DBN-GRU model provides a comprehensive method for stimulating complex information connections in network data. TensorFlow and Keras libraries are utilised to create and train the neural network model. Three crucial steps in the training process are progressive backpropagation, hyperparameter adjustment, and regularisation. A 98.92% accuracy rate in categorising incoming data as either normal or abnormal network activity is achieved by the model, according to an independent dataset evaluation. Compared to baseline methods like Random Forest, SVM, and GNB, this is a significantly better approach. The results of the study demonstrate how well the suggested strategy works to identify intrusions and enhance network protection. Python is used as a software tool throughout the implementation, which improves the approaches' dependability and efficiency. This work offers a helpful tool for enhancing network security through the creation of distinctive hybrid architecture by accurately detecting and classifying intrusion attempts. The approach, outcomes, and conclusions of the study offer significant insights into how intrusion detection systems and machine learning can be used in IT systems to secure data and maintain information security.
| Item Type: | Thesis (Masters) |
|---|---|
| Supervisors: | Name Email Salahuddin, Jawad 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 > T Technology (General) > Information Technology 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: | 21 Apr 2025 13:29 |
| Last Modified: | 21 Apr 2025 13:29 |
| URI: | https://norma.ncirl.ie/id/eprint/7453 |
Actions (login required)
 |
View Item |