Adaptation of Machine Learning in Fog Computing: An Analytical Approach

Abstract

With the growth of the world, IoT and Cloud computing is becoming very highly sophisticated technological application widely used in various fields, like medical science, transportation, industry, monitoring environment, smart city, gaming, home automation, security, etc. Whenever we are talking about IoT and cloud computing there is an extension of this paradigm to the edge of the network namely Fog computing. Fog computing plays a crucial role in providing faster response time and improving network traffic by reducing latency time while executing any kind of task. The main purpose of fog computing is to reduce the burden of the cloud with low latency in a distributed manner. However, sometimes due to some issues, Fog computing has failed to provide adequate and accurate results that reduce effectiveness and quality in performance operations. For this reason, Machine learning (ML) can be used to increase the speed and transmission processes of data through Fog nodes. It helps in improving the architectural series of Fog nodes through real-time processing and communication processes that can be developed according to the user's expectations. This paper tried to present a comprehensive review of the role of Machine Learning in Fog computing by exploring the latest adaptation of ML techniques in some key aspects of Fog (Resource management, security, and Computational enhancement).