DDoS Attack Detection using Multivariate Correlation Analysis and Clustering

Author(s):

Ravi Pawar , Sinhgad Academy of Engineering,Kondhwa; Devta Anekar, Sinhgad Academy of Engineering,Kondhwa; Amarita Mishra, Sinhgad Academy of Engineering,Kondhwa; Mansi Pawar, Sinhgad Academy of Engineering,Kondhwa; Ravi Pawar, Sinhgad Academy of Engineering,Kondhwa

Keywords:

ETL, Database, Reporting System, DDoS, Analytics

Abstract:

An assault on a network that floods it with numerous requests that regular network traffic is either slowed or completely interrupted. Unlike a virus or worm which can cause severe damage to ETL process, Database and Reporting System as well. The reliability and availability of network services are being threatened by the growing number of Denial-of-Service (DoS) attacks on web server. Effective mechanism for DoS attack detection is demanded. Such detection system needs to implement which will capable to provide analytical data using statistical analysis. Different systems were proposed for detection DoS attacks using machine learning, statistical analysis, data mining, etc. The proposed system is enhancement of earlier one in which k- means clustering technique is applied over a training data samples so that it can categorize the samples into different clusters and then it applies statistical analysis methods to find the correlation between features to reap statistical mathematical information like standard deviation, mean and covariance matrix. While applying multivariate correlation analytics on each cluster based on threshold value of clustered data set will help to get profile parameters according to related cluster and will get to know sharp boundary of characterizing a sample packet. Ultimately, this will reduce false positive rate and will beef up the accuracy.

Other Details:

Manuscript Id	:	IJSTEV2I12015
Published in	:	Volume : 2, Issue : 12
Publication Date	:	01/07/2016
Page(s)	:	5-9

Download Article