Fog Data Analytics: Systematic Computational Classification and Procedural Paradigm

Abstract

The unprecedented proliferation and acceptance of sensor-based gadgets has given rise to technological advancements and innovations in all of the interconnected technologies related to it. The transactional data that is generated from this level of connectivity is termed as Big Data and is continuously collected and delivered in endless streams between the networked entities. Big data contains insightful information and knowledge that can be exploited by businesses by mining and analytics. But the sheer volume of data brings with it a lot of challenges in terms of efficient data storage and effective data analytics on fog and cloud computing platforms generally and specifically at the edge of the cloud on fog devices. Extensive research has been carried out to address the challenges and realize the benefits of fog data analytics. In this chapter, we discuss the characteristics of attributes and analyze the algorithmic complexities in fog data analytics. A systematic computational classification is generated to develop a paradigmatic procedure for fog data analytics to process, store and analyze data efficiently and effectively and develop an ideal platform for proliferating sensor-based devices and services on Internet of things. We present few case studies that benefit from the proposed model.