Application of Machine Intelligence and Data Science for Intelligent IoT

Abstract

Applications for the Internet of Things have evolved in excessive quantities, producing a vast volume of data needed for the intelligent transformation of data. The numerous IoT frameworks (i.e. edge, fog, and cloud) and the shortcomings for the protocols of the IoT application level in the transmission/receipt of messages are, however, the obstacles to the development of acute IoT approaches. The challenges preclude existing acute IoT frameworks that learn from other IoT applications in an adaptive manner. In this article, we critically discuss how machine learning research processes the IoT-generated data and highlight the existing difficulties of fostering intelligent approaches for the IoT world. In addition, the authors suggest a foundation to allow IoT frameworks to learn appropriately from various other IoT frameworks and provide a chronology on how to apply the approach to specific literature surveys. Finally, we answer the essential variables which have an effect on potential IoT intelligent frameworks.