Unsupervised and Semi-supervised Machine Learning Algorithms for Cognitive IoT Systems
Pethuru Raj, Anupama C. Raman, Harihara Subramanian · Auerbach Publications eBooks · 2022
Machine learning (ML) algorithms are gaining hitherto unheard prominence across industry verticals. Several prediction problems have solutions with the straightforward application of ML algorithms. For the next era of knowledge, the contributions of the machine and deep learning algorithms are becoming extraordinary. Especially ML algorithms come in handy in realizing intelligent business workloads and IT services. Further on, with the steady accumulation of IoT devices, machines, instruments, and equipment in prominent places, the challenge is to enable them to be intelligent in their operations. Leveraging prime methods, the innovative embedding of ML libraries and frameworks into the application to express and expose them as data-crunching cognitive systems to the outside world is trending. This methodical transition helps consumer electronics, personalized devices, and industry machineries to interact and collaborate to fulfil peoples’ everyday needs (time-centric, context-sensitive, intellectual, information and service). In the previous chapter, we discussed supervised ML algorithms and their unique use cases. In this chapter, we are to focus on unsupervised and semi-supervised ML algorithms along with their special applications.