Learning from Privacy Preserved Encrypted Data on Cloud Through Supervised and Unsupervised Machine Learning

A Review of Machine Learning Algorithms for Cloud Computing Security

“…El-Baghdadi et al [64] used one of the ML emergent algorithms, DRLCS, to solve the problem of resource scheduling in CC. Ahmad et al [15] discussed the lack of interest of the data owner that still exists due to data insecurity by a third party that stores, manage, and processes the data. In Reference [64], the advantage is to improve the data analysis and the disadvantage is the challenge of selecting the appropriate resource scheduling algorithm for a specific workload.…”

“…In Reference [64], the advantage is to improve the data analysis and the disadvantage is the challenge of selecting the appropriate resource scheduling algorithm for a specific workload. In Reference [15], the advantage is cloud workload protection and the disadvantage is to dedicate and specialized client-server applications for proper functionality. Zamzam et al [69] discussed resource management using ML in MEC to solve the issue and improve performance.…”

“…Khan et al [15] discussed an algorithm to solve security issues to improve the performance of the cloud system. The lack of interest of information still exists because of information instability by the outsider, which stores, manages, and forms the information.…”

“…This innovation empowered hybrid applications to be designed with code move that sudden spike in demand for various working frameworks, (for example, Android or Windows), which decreases the amount of work required from engineers, as a similar code is executed on a cell phone and in the cloud. The authors in [15] investigated both supervised and unsupervised ML capabilities through ANNs over encrypted data from a semantically secure cryptosystem based on homomorphic properties. The lack of interest of information owners still exists because of information instability by the outsider, which stores, manages, and forms the information.…”

“…Ahmad et al [15] performed a survey of ML techniques for a secure Cloud. The lack of interest of data owners still exists due to data insecurity by third-parties that store, manage, and processes the data.…”