Managing Smart Cities and Communities Using Machine Learning/Deep Learning Approaches based on Emerging Technology Trends: An Extensive Literature

Abstract

The aim of managing smart cities and societies is to optimize the use of limited resources while improving residents' quality of life. To achieve sustainable urban living, smart cities implement new technologies like the Internet of Things (IoT), Internet of Drones (IoD), and Internet of Vehicles (IoV). The data generated by these technologies is analyzed to gain insights that enhance the efficiency and effectiveness of smart communities. Common applications in smart cities include smart traffic management, energy management, city surveillance, smart buildings, and healthcare monitoring. Additionally, Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) show significant potential for automating various functions within smart cities. This research explores various challenges and future research directions where these technologies can advance the smart city concept. Specifically, it aims to provide a better understanding of (1) the fundamentals of managing smart cities and societies, (2) recent advancements in the field, (3) the advantages and limitations of current methods, and (4) areas needing further exploration. Findings indicate that Conventional Neural Networks (CNN) and Long Short-Term Memory (LSTM) are the most commonly utilized ML methods in the literature. Most studies focus on power and energy management within smart cities, often concentrating on a single parameter, with accuracy being the primary focus. Additionally, Python is the most widely used programming language, appearing in 69.8% of the reviewed papers. This paper provides a literature review on the smart city concept, sustainability initiatives within smart cities, their functional aspects, and a survey of the applications of ML and DL in this context.