Machine learning algorithms for detecting DDoS attack in cloud computing

Abstract

The Distributed Denial of Service DDoS attack is a kind of intrusion in cloud computing environment that severely affects the end user by injecting illegitimate packets of data into internet traffic without the knowledge of the clients It is a serious problem in cloud computing because the detection and mitigation of intrusion is a challenging task that will affect the functionality of the entire architecture. Numerous cyber security measures have been carried out to protect the server from attackers or hackers The traditional cyber security methods failed to protect the server against several external unauthorized traffics It is important to develop Intrusion Detection System IDS in loT architecture Detailed literature reviews are carried out to investigate various machine learning techniques neural network models and optimization algorithms are aimed to identify the gap problems and to then develop machine learning algorithms to detect the intrusion accurately and effectively The data mining algorithms such as C4 5 SVM and KNN are developed and their performances are investigated in detecting DDoS attack The performance of the developed classifier algorithms is compared with other classifier algorithms such as Random Forest Naive Bayes and CART Based on the result analysis made the SVM based DDoS attack detection model outperformed all other algorithms but it suffers to handle large dataset newline