AN APPROACH FOR SECURING BIG DATA ENVIRONMENT USING MACHINE LEARNING FOR DYNAMIC PROCESSING OF PORTABLE EXECUTABLE IN NETWORK

Cyberattack landscape is rising at a pace which was never before. Scale of mobile and network attacks have become a common place due to increased online activities. In such a scenario, it is imperative for organizations to detect these attacks well in advance so that compromising of sensitive information is prevented. In the recent past, machine learning techniques are being extensively used for detection as well as prevention of these attacks. In this paper, a model is designed for detecting malware in virtualized environment. This is done by extracting features from Portable Executable (PE) file from data nodes and using various combinations of feature preprocessing and classification models. Among the four combinations, performance of Haar Wavelet Transform (HWT) and Naïve Base Classifier was found more promising.