Detection of DDoS attacks using RNN-LSTM and Hybrid model ensemble

Kona, Siva Sarat (2020) Detection of DDoS attacks using RNN-LSTM and Hybrid model ensemble. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration manual) Download (1MB) | Preview

Abstract

The primary concern in the industry is cyber attacks. Among all, DDoS attacks are at the top of the list. The rapid increase in cloud migration also increases the scope of attacks. These DDoS attacks are of different types like denial of service, distributed denial of service, Slowloris, and so on. There are many implementations to detect the attacks. There are different types of detection systems and using popular machine learning techniques. A lot of research is going under the improvisation of machine learning techniques. The existing implementations are proved to predict better results with classifiers like Decision Trees, Support Vector Machine(SVM), Logistic Regression, and Neural networks. Also, many types of research proved to be more efficient by combining the algorithms to achieve high accuracy. Usually, network data is immense. So, generating and maintaining a hybrid model requires high execution time and more resources. My model is a solution with the hybrid implementation of the model using ensembling. Recurrent neural networks are used in weather, share prices, e-commerce and typing prediction. Many big players in this industry were adopting to this prediction model. This widely used time series analysis algorithm is used to predict the anomaly within the dataset. Based on the prediction period, the data is sent to a hybrid model to detect the attack record. This hybrid model is built on the high accurate prediction model "Random Forest" along with high customizable algorithm "Neural Network". With this implementation, I can achieve an accuracy of 95.2% and 83%, respectively. The ensemble model with these two algorithms chooses the best model based on model voting. The final model built with an accuracy equal to the Random forest.

Item Type:	Thesis (Masters)
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 T Technology > T Technology (General) > Information Technology > Computer software 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 Cyber Security
Depositing User:	Caoimhe Ní Mhaicín
Date Deposited:	03 Apr 2020 15:39
Last Modified:	03 Apr 2020 15:39
URI:	https://norma.ncirl.ie/id/eprint/4180

Actions (login required)

View Item