MACHINE LEARNING ALGORITHMS FOR CLOUD COMPUTING SECRECY: A STATE OF THE ART SURVEY

Abstract

On-demand accessibility of network resources, in particular data storage and processing power, is provided by Cloud Computing (CC) without special and direct management by users. CC has recently emerged as a collection of public and private data centres that offer a single platform across the Internet to the customer. Edge computing is an emerging computing paradigm that takes end-users closer to computing and data storage to increase response times and spare transmission power. Mobile CC (MCC) transmits apps to cell phones using distributed computing. CC and edge computing, however, have security problems, including client vulnerability and association recognition, which hinder the rapid adoption of computing models. The study of computer algorithms that naturally develop through practise is machine learning (ML). We present an overview of CC security risks, challenges, and solutions using one or more ML algorithms in this review article. We study various ML algorithms, including supervised, unsupervised, semi-supervised, and reinforcement learning, that are used to solve cloud security issues. Then, based on their characteristics, benefits, and drawbacks, we compare the efficiency of each technique. In addition, to secure CC models, we enlist potential research directions.