Implementation of Machine Learning Techniques for Risks Evaluation in Cloud and Cybersecurity

Abstract

Cloud computing has emerged as an essential element of the modern and future industry. Various corporates utilizes the capabilities of cloud computing services. There is skeptism and fear regarding application of cloud services that remains an open challenge, as it gaining and growing popularity amongst many business entities around globally. Many challenges are determined and found out research; majorly pertaining to security or protection. Since its inceptions Security risks concerning to cloud computing has invited large portion of attention. Cloud computing services and their providers are always on the lookout for new and improved security methods and solutions. Many service providers in field of cloud computing exist with their services in cloud domains. Such services comes with many features, specifications, along with techniques of acquiring security measures. Methodologies acquired and adopted by many service players to attain security is different in nature. Depending on his need and quality of security received from service providers. A user can choose a particular service. For studying a specific service that depends on its many security features is a dominant issue. In order to build a comprehensive risk assessment methodology, an extensive literature review was conducted to identify all risk factors that can affect cloud computing adoption. In this context various risk factors were identified. After feature selection methods and identification of risk factors, utilized to select most effective features. Then machine learning techniques are used as an efficient technique to analyze hazard in an environment of cloud computing. From all the partitioning strategies tested, the results showed that dividing the dataset into 96% and 4% yielded the best results. The Decision Tree Classifier method also performed the best across all the datasets.