Study on Web DDOS Attacks Detection Using Multinomial Classifier

Detecting DDoS attacks at application layer is quite challenging research problem. The recent methods are suffered from the poor accuracy performance of DDoS attack detection at application layer. In this paper, to mitigate current problems, classifier based system is proposed in which packets are captures, extraction of important fields those are required for detection and then apply classifier to detection of attack. In this paper, we studied and discussed the algorithm of Naive Bayes Multinomial for testing and training. The performance of this approach is compared with other existing classifiers into the terms of accuracy, true positive & false positive rates. The outcome of this paper is current method limitations and scope of improvement depicted from overall study and analysis. Additionally, the aim of this paper is to identify the research gap and limitations of studied method with review of previous methods.