Predicting Application Layer Ddos Attacks Using Machine Learning Algorithms

S. Umarani, D. Sharmila

Zenodo (CERN European Organization for Nuclear Research) · 2015 · 27 citations · 0 references

DOIFull text

Open access

Concepts

Abstract

A Distributed Denial of Service (DDoS) attack is a<br> major threat to cyber security. It originates from the network layer or<br> the application layer of compromised/attacker systems which are<br> connected to the network. The impact of this attack ranges from the<br> simple inconvenience to use a particular service to causing major<br> failures at the targeted server. When there is heavy traffic flow to a<br> target server, it is necessary to classify the legitimate access and<br> attacks. In this paper, a novel method is proposed to detect DDoS<br> attacks from the traces of traffic flow. An access matrix is created<br> from the traces. As the access matrix is multi dimensional, Principle<br> Component Analysis (PCA) is used to reduce the attributes used for<br> detection. Two classifiers Naive Bayes and K-Nearest neighborhood<br> are used to classify the traffic as normal or abnormal. The<br> performance of the classifier with PCA selected attributes and actual<br> attributes of access matrix is compared by the detection rate and<br> False Positive Rate (FPR).