Taxonomy and Implications of Machine Learning for Internet of Things: Qualities, Uses and Algorithms

The future Internet of Things will have profound economic, commercial and societal implications. IoT nodes are often resource restricted, making them attractive targets for cyber assaults. In this sense, classic cryptographic algorithms have been extensively used to solve security and privacy challenges in IoT networks. However, due to the unique properties of IoT nodes, traditional technologies cannot fully secure IoT networks. Machine Learning (ML) and Deep Learning (DL) approaches may help solve many security issues in IoT devices and networks. This article discusses how many firms employ machine learning approaches to analyse data and uncover useful data forms and key components of the many scenarios of IoT with machine learning algorithms, as well as the comparison of different machine learning algorithms with IoT.