Machine Learning and Deep Learning for Intelligent and Smart Applications

Abstract

Machine learning (ML) and/or deep learning (DL) are rapidly changing various technologies used by us in our daily lives. To make specific decisions, ML requires a rational amount of data. ML as well as DL are a subset of artificial intelligence and comprise more sophisticated methods and technologies that allow computers to understand the information and deliver smart and intelligent outputs. The science of having computers to function without being programmed directly is ML. ML is a concept that has recently become a subject of focus, although it has been developed since several decades, primarily because of the advancements in DL and artificial neural networks (ANNs). Primary ML methods have made it possible for people to step dramatically forward in cognitive areas such as natural language processing (NLP), image recognition, and text analysis. DL is a growing paradigm of ML that makes use of multi-layered ANN in applications such as speech recognition, classification, language translation, as well as further modern breakthroughs that are in the news to provide high precision. The uniqueness and benefit of DL is that without addition of conventional hand-coded programmes or rules, it can mechanically extract or translate or learn various features like video, text, or images from data sets. The main objective of this chapter is to compare the most commonly used DL and ML techniques used in smart grid applications, like convulational neural network (CNN), support vector machine, decision tree, neural network, and descriptive discriminant.