ONIETAN, IYANU-OLUWA CHRISTOPHER and Covenant University, Theses (2023) NETWORK INTRUSION DETECTION MODEL USING ENSEMBLE-BASED LEARNING. Masters thesis, COVENANT UNIVERSITY.
![[img]](http://eprints.covenantuniversity.edu.ng/style/images/fileicons/application_pdf.png) |
PDF
Download (139kB) |
Abstract
The interconnectedness of devices, technologies, networks and the services they provide has continued to increase. This has also resulted in increased cases of cyber threats and intrusions. detection has become a major concern for organizations. Intrusion detection system is one way to address the issue of intrusions and anomaly network traffic. Existing machine learning algorithms has performed well on intrusion detection however, the issues of high false positive rates as well as low accuracy still persists. This is largely due to fact that individual models are not able to efficiently detect previously unknown intrusions on their own. Other the hand, ensemble models have proven to be more efficient in identifying intrusions and anomalies in networks since they combine the predictive powers of several base models. However, the efficiency of ensemble models has not been sufficiently considered where imbalanced datasets are involved. This study therefore proposes and investigates the performance of various ensemble models when applied to conspicuously imbalanced datasets. Two largely imbalanced datasets were acquired namely IDT2 and CICIDS2017. Additional datasets were generated from each of the acquired datasets using SMOTE and SMUTE, oversampling and under-sampling techniques respectively. In order to investigate the performance of ensemble models, three ensemble models were constructed namely Bagging, Majority voting and Stacking. The performance of each model was effectively determined and compared.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Network; Intrusion Detection Systems, Intrusion Prevention Systems, Machine Learning; Cybersecurity; Ensemble Models. |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | UNSPECIFIED |
| Depositing User: | AKINWUMI |
| Date Deposited: | 03 Oct 2023 11:43 |
| Last Modified: | 03 Oct 2023 11:43 |
| URI: | http://eprints.covenantuniversity.edu.ng/id/eprint/17328 |
Actions (login required)
 |
View Item |
