Data mining techniques for IoT and big data — A survey

Abstract

Data Mining is the discovery of “models” of data. Data dredging is a process of derogatory referring to attempts for extracting information that was not supported by the data. Today, data mining is more similar to machine learning and most techniques uses algorithms from machine learning in order to discover unusual events hidden within the large amounts of data. The recent advancements in communication technology, people and the things are becoming increasingly interconnected. The availability of the Internet makes it possible to connect various devices that can communicate with each other and share data. The Internet of Things (IoT) is a new concept that allows users to connect various sensors and smart devices to collect real-time data from the environment. Big Data is a vast amount of data collected from IoT environment and it applies to information that can't be processed or analyzed using traditional tools. Every organization is facing more and more challenges to access a wealth of information and how to get values out of large variety of data. As creation of data is much easier than analyzing it, there is a need for Novel approaches in data mining techniques to deal with huge data. From the perspective of software, the traditional mining algorithms are applicable only for small scale IoT data. This paper first focuses on a review of existing techniques and data mining algorithms that are used to process massive data of IoT and also the limitations are discussed. Second, the work and advancements related to data mining algorithms that are implemented with Hadoop and Spark technology are presented. Third, Hybrid data mining algorithms using MapReduce framework are reviewed. Finally, open research challenges and issues are presented as a conclusion.