A Robust DNS Flood Attack Detection with a Hybrid Deeper Learning Model. (May 2022)

• Record Type: Journal Article
• Authors: KASIM, Ömer
• Is Part Of: Computers & electrical engineering. Volume 100(2022)
• Publication Date: 2022-05
• Subjects: Domain Name System Flood Attack Detection -- Deeper Convolutional Neural Network -- Long Short Term Memory Model -- Network Security; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.107883
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: In this article, an innovative hybrid deep learning model which consists of convolutional neural network architecture that improves with long short term memory cells is used to recognize domain name system flood attacks. The proposed model provides a comprehensive solution for direct detection of domain system flood attacks without label coding, normalization and feature selection in CICIDS dataset derived from real world data. The performance metrics is evaluated through a comprehensive performance comparison of selected machine learning, shallow neural network and deep learning classifiers. Although there is no normalization, label encoding and feature removal, very low false alarms and significantly high detection accuracy (99.87%) have been achieved. The deep learning architecture, including the proposed Long Short Term Memory cell, has contributed to the detection of domain name system flood attacks with a low false-positive diagnosis rate compared to machine learning and shallow neural networks.