Research Area: Cloud Security
Recent days have seen an apparent shift in most of the organizations moving towards using cloud environment and various cloud-based services. In order to protect and safeguard the transactions made by organizations over cloud environment, it is highly essential to provide a secure and robust environmental solution across cloud space. Existing approaches such as linear regression and support vector machine have been tried to promote cyber-security in the market by performing static verification of cloud user behaviour in order to identify pre-defined threats. Due to their static nature, these security solutions are restricted in their functionality. When it comes to access control, the decision making involves performing a permit or block operation. Also, the earlier methods face difficulties in terms of data protection over the endpoints which are not managed by the cloud. In order to solve the above-said problems, this paper is focused on designing a novel security solution for cloud applications using machine learning (ML) approaches. The main objective of this paper is to shape the future generation of cloud security using one of the ML algorithms such as convolution neural network because CNN can provide automatic and responsive approaches to enhance security in cloud environment. Instead of focusing only on detecting and identifying sensitive data patterns, ML can provide solutions which incorporate holistic algorithms for secure enterprise data throughout all the cloud applications. The proposed ML algorithm is experimented, results are verified and performance is evaluated by comparing with the existing approaches.
Keywords:
Machine learning
Cloud security
Malicious threats
Author(s) Name: E. K. Subramanian & Latha Tamilselvan
Journal name: Service Oriented Computing and Applications
Conferrence name:
Publisher name: Springer
DOI: 10.1007/s11761-019-00270-0
Volume Information: volume 13, pages 237–249 (2019)
