• Dhanashri A. Hase
• Shrinivas K. Sonkar

As a prime strategy to make certain the safety of IT infrastructure, anomaly detection plays a greater vital position in cloud computing platform which hosts the entire applications and information. On pinnacle of the traditional Markov chain version, we proposed on this paper possible multi-order Markov chain based totally framework for anomaly detection. In this approach, both the high-order Markov chain and multivariate time collection are followed to compose a scheme described in algorithms at the side of the schooling manner in the form of statistical mastering framework. To cut down time and area complexity, the algorithms are designed and carried out with non-zero cost table and logarithm values in initial and transition matrices. For validation, the collection of system calls and the corresponding return values are extracted from conventional Defense Advanced Research Projects Agency (DARPA) intrusion detection assessment facts set to shape a -dimensional take a look at input set.In this paper using spatio-temporal for they have been analyzed for various types of human activity including execution of political actions, disaster management, crime prevention, emergency services, etc . Such data can be regarded as a collection of spatio-temporal documents.

Kth-order Markov chain, multivariate time series, anomalydetection, statistical learning.

"Anomaly Detection Using Statistical Learning In Cloud Server System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.452-458, June-2019, Available :http://www.jetir.org/papers/JETIR1906B72.pdf

"Anomaly Detection Using Statistical Learning In Cloud Server System", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp452-458, June-2019, Available at : http://www.jetir.org/papers/JETIR1906B72.pdf