Cloud industry has reached a critical mass in the past few years, with many cloud service providers fielding competing services. Despite the competition, some of the security mechanisms offered by the services to be similar, indicating that the cloud industry has established several “best-practices,” while other security mechanisms vary widely, indicating that there is also still room for innovation and experimentation. The cloud industry had grown in the fast few years, with so many providers focusing on cloud services. Besides huge competition the security mechanisms that is provided by all these vendors had shown many good practices but that paves a way for many new innovative experiments. This paper mainly focusses on improving the security mechanism against DDOS attacks. With the existing system we are not able to predict the magnitude of DDOS attack as the causes vary across different situation. So, resolving this security issue becomes much more complex in real time situation. One important reason for DDOS attacks can be because of fake users creating spoofed request. Apart from that there are also additional attacks which are made within cloud environment and outside cloud environment, so security mechanisms must be tightened. There is also some hidden pattern which prevails on user surfing through websites based on their frequency and content visited which is also required to establish furthermore security based on user behavior. The aim of the paper is to predict the magnitude of DDOS attack which is bonded with a two-fold solution 1. Capturing trust rating for a user visiting a website considering his frequency and website safety ranking based on a Scaled User Rating Algorithm. 2. Considering parameters that helps in figuring DDOS attack pattern based on both internal and external attacks within the cloud environment. The aims defined in this paper help us in figuring out a malicious behavior of user based on his surfing pattern and contents that he had referred, which in turn help us in expecting a possible DDOS attack. In addition to that we are also trying to find possible parameters that could be a reason for DDOS attack analyzing threats that had happened within and across cloud environments in the past.

Cloud Computing; Data Privacy; Data Protection; Security; Virtualization; Monitoring; Deep Learning; Predictive Analytics; Scaled User Rating

[1] Sakyajit Bhattacharya, Tridib Mukherjee, and KoustuvDasgupta, “CloudRank: A Statistical Modelling Framework for characterizing user behaviour towards targeted Cloud Management” IEEE Network Operations and Management Symposium, 2014.
[2] LI Jun-Jian, Li-Qin, “User’s Behavior Trust Evaluate Algorithm Based OnCloud Model” IEEE Fifth International Conference on Instrumentation and Measurement, Computer, Communication and Control, 2015
[3] Xiaoming Ye, Xingshu Chen, Haizhou Wang, XuemeiZeng, Guolin Shao, Xueyuan Yin, and Chun Xu, “An Anomalous Behavior Detection Model in Cloud Computing”Special Issue On Information Security ,Volume 21, Number 3, June 2016
[4] Xin Lu, Cheng du, China; Yue Xu, Cheng du, China “An User Behavior Credibility Authentication Modelin Cloud Computing Environment”. IEEE International Conference on Information Technology and Electronic Commerce, 2014.
[5] Mahesh Babu, Mary SairaBhanu, “Analyzing User Behavior Using KeyStroke Dynamicsto Protect Cloud from Malicious Insiders” IEEE International Conference on Cloud Computing in Emerging Markets, 2014
[6] Lucky Nkosi, Paul TarwireyiMathew O Adigun, “Detecting a Malicious Insider in the CloudEnvironment Using Sequential Rule Mining”, IEEE International Conference on Adaptive Science and Technology, 2014
[7] Ngugi, Benjamin, Beverly K. Kahn, and Marilyn Tremaine. "Typing biometrics: impact of human learning on performance quality. " Journal of Data and Information Quality (JDIQ) 2. 2 (2011): 11
[8] Teh, Pin Shen, Andrew BengJin Teoh, and Shigang Yue. "A survey of keystroke dynamics biometrics. " The Scientific World Journal 2013 (2013).
[9] LEE, K., CAVERLEE, J., AND WEBB, S. Uncovering social spammers: social honeypots + machine learning. In ACM SIGIR: Proceeding of the international conference on Research and development in Information Retrieval (2010)
[10] R.Piplode, P. Sharma and U.K. Singh, "Study of Threats, Risk and Challenges in Cloud Computing", International Journal of Scientific Research in Computer Science and Engineering, Vol.1, Issue.1, pp.26-30, 2013
[11] P. Rutravigneshwaran, "A Study of Intrusion Detection System using Efficient Data Mining Techniques", International Journal of Scientific Research in Network Security and Communication, Vol.5, Issue.6, pp.5-8, 2017