Decision Trees to Detect Malware in a Cloud Computing Environment

Abstract

Cloud computing is the most demanding technology for efficient storing of information and provide facilities online. As the data is stored in huge amount, it is very much prone to cyber-related attacks. Using this increasing technology to protect storage information which is computer based from malware attacks and show up several rewards along with traditional detection schemes. Because cloud-based protected products are portable, they may be used on any computer-based system, including Cyber Physical Systems (CPS), personal devices such as Internet of Things (IoT) devices, laptop, desktop and mobile devices. The malicious software is computer software that is designed to launch exploitable assaults against computer systems that gains unauthorized access to the system which confidentiality, integrity and authentication (CIA) requirement triad agreement has been distributed, then the whole organization will be in trouble is generally called as malware. To detect various forms of malware attacks, the effective malware detection (MD) technique is proposed in the work in a cloud based data storage environment. The proposed MD technique will be learnt with the dataset that contains characteristics on different virtual machines (VM) which identify distinctive MD features in an efficient manner. The decision tree (DT) is then given selected features. For identifying malware and benign samples, a DT employs learning and rule-based (LRB) agents. The proposed technique can identify any software with high detection rate and accuracy. It will be characterized based on the features, and falls on two buckets - malware and benign.