Tackling the Cloud Adoption Dilemma - A User Centric Concept to Control Cloud Migration Processes by Using Machine Learning Technologies

• Published in: ARES 2016 : proceedings : 2016 11th International Conference on Availability, Reliability and Security : Salzburg, Austria, 31 August - 2 September 2016 pp. 776 - 785
• Main Authors: Diener, Michael, Blessing, Leopold, Rappel, Nina
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2016
• Subjects: cloud adoption; Cloud computing; cloud migration; cloud security; Companies; Context; data classification; Encryption; machine learning; Machine learning algorithms; naive bayes classifier; Sensitivity; supervised learning; user centric tool
• DOI: 10.1109/ARES.2016.39

Research studies have shown that especially enterprises in European countries are afraid of losing outsourced data or unauthorized access. Despite various existing cloud security mechanisms companies are currently hesitating to adopt cloud resources. This phenomenon is also known as cloud adoption dilemma. We think that data classification is a promising technique that should be considered in the context of cloud security, supporting cloud migration processes. By using classification techniques enterprises are able to control which documents are suited for Cloud Computing and which cloud service providers are sufficient for protecting sensitive documents. In this work we present an efficient concept that involves enterprises' employees and authorities, making it possible to apply powerful security policies in a simple way. We make use of a well-established machine learning algorithm in our developed tool, identifying security levels for different types of documents. Thus, cloud migration processes can become more transparent and enterprises obtain the ability to discuss more openly about adopting innovative cloud services.