Systematic Review of Machine Learning and Deep Learning Applications in the Development of Smart Homes Using IoT (#2032)

Authors

Ocaña Velásquez, Jesus Daniel

Castro García, José Heiner

Miranda Saldaña, Rodolfo Junior

Abstract

The advancement of technology has led to a remarkable development in smart homes, where remote and automated management of connected devices transforms the user experience. This article systematically reviews the applications of Machine Learning and Deep Learning in the development of smart homes through the Internet of Things (IoT) technology. The objective of this research is to carry out a systematic review on the application of Machine Learning and Deep Learning in smart homes connected to IoT, focusing on energy efficiency, security and comfort, as well as identifying trends, challenges and opportunities for improvement to serve as a guide for researchers and developers. The PRISMA method was used to gather 68 relevant articles. The results show that Machine Learning and Deep Learning play a fundamental role in this field, with a greater number of investigations carried out in China and India. The most common methods in Machine Learning are Random Forest and Decision Trees, while in Deep Learning LSTM and CNN stand out. It is concluded that Machine Learning and Deep Learning are essential to improve security and Quality of Service by identifying and solving problems in advance. Deep Learning, in particular, improves surveillance and motion detection, and its combination with Machine Learning promises to transform monitoring, creating more integrated and efficient management systems.