An Overview of Concepts, Applications, Difficulties, Unresolved Issues in Fog Computing and Machine Learning
Oscar Jayanagara, Dewi Sri Surya Wuisan
Abstract
Numerous fog computing apps and services are emerging as a result of the large volumes of data produced by systems based on fog computing. Additionally, the crucial field of machine learning (ML), which has made significant advancements in several academic areas, incorporating speech recognition, robotics, and neuromorphic computing, in addition to computer graphics and natural language processing (NLP).. The aim of current research provided insight into providing a list of the fog computing-related ML operations. The security, capacity, and latency standards for networks must be met by many IoT applications. Cloud computing does not, however, satisfy these needs. Today's technology can satisfy these objectives, and edge computing is one such option. The model enables traffic analysis and sensor data analysis. The management of resources, accuracy, and security are three areas of fog computing that we highlight in this thorough assessment of the most recent advancements in ML approaches. Additionally highlighted is the function of ML in edge computing. Additional viewpoints on the ML domain are presented, including those on the different kinds of application support, techniques, and datasets. Finally, open questions and research difficulties are highlighted.