A survey of data mining methodologies in the environment of IoT and its variants

In today's world, the number and variety of objects that are connected to the Internet of Things (IoT) infrastructure is increasing exponentially. Consequently, data produced by these devices also increases in the exponential rate. These massive data needs to be analyzed for extraction of knowledge and information using suitable data mining tools. This analysis plays a very important role in healthcare, farming, enterprises etc. for taking appropriate decisions. As it is well known that the IoT system suffers from many limitations like resource and energy constraint, heterogeneity of devices and data, scalability, so the implementation of data mining tools in such infrastructure becomes a challenge. Data mining techniques needs to be adapted for such infrastructure. Many investigators have put their endeavors in the IoT data mining field. In this paper, a survey on those investigations has been discussed in detail. Recently, several new architectures of IoT have been investigated which are capable of overcoming the standard conventional limitations of IoT. One of the most prominent architectures in such direction is Software Defined Network (SDN) based IoT. An insignificant number of works have been done on Data mining applications in SDN based IoT platforms. This paper presents detailed study and critical analysis of past works related to IoT and SDN based IoT data mining. Prominent research gaps have been investigated and outlined. Future research directions in the field have been also outlined.