Machine learning algorithms for distributed operations in internet of things IoT

Abstract

Generally, the things that have the great role in facilitating the emergence of internet-connected sensory devices can be embodied in the developments that happen in the sphere of software, hardware, and communication technologies. The internet-connected sensory devices present perceptions and measurements of data from the real world.  It is suggested that nearly through 2020, the total use of internet-connected devices may reach to 25 to 50 billion. Actually, the relation between technologies and the volume of data being published is kept in one line. That is, if there is growth in the technologies, the volume of the data will be increased. Such technology, i.e. internet-connected devices, can be called as Internet of Things (IoT). Its role is to connect the real world with the cyber one. Furthermore, generating great data with velocity as its main characteristic will help in increasing the volume of IoT.  To develop smart IoT applications, one can use such intelligent processing and analyzing such big data. In this paper, we tend to study the impact of implementing machine learning (ML) algorithms and methods and their efficiency in the IoT domain. As well as explore how these algorithms help in founding efficient backbone solutions to analyze and estimate the huge amounts of data that are expected to arise in the coming few years due to the rapid growth on demands for IoT based applications.