RESPOND OF SMART CITIES TO PANDEMIC COVID-19

Abstract

Over the past several years, smart cities have gained widespread recognition worldwide because they facilitate a better life for people and bring substantial changes in society; more specifically, smart cities impact the lives of citizens amid the pandemic COVID-19. The COVID-19 epidemic has affected many elements of life, including economics, education, healthcare, politics, and financial planning. However, case studies of 120 cities have been examined concerning their responses to pandemics, the pillars of smart cities, and the number of COVID cases. The study investigates how pandemic preparedness in smart cities could be improved by stopping the spread of the virus, finding treatment (vaccine), and remotely providing education and business continuity. Smart city solutions are desperately needed to deal with the multiple challenges of controlling the COVID-19 epidemic. One of the approaches is to forecast the spread of the virus and its effects on various areas of urban life using machine-learning models. We can test the effectiveness of different regression models in both scenarios using a dataset that contains relevant variables for the COVID-19 and smart city pillars. The prognostic accuracy of the Elastic Regression, Ridge Regression, Lasso Regression, Random Forest Regression, and MLP Regression models for the prediction of COVID-19 infection rates based on smart city characteristics will be assessed in this way. The Random Forest Model outperformed Ridge Regression after adjusting model parameters and hidden layers (R-squared score 0.88 and 0.85, respectively). The MAE and MSE scores obtained using the Random Forest Regression technique are 0.009692 and 0.000261, respectively. On the other hand, the unsupervised ML Elbow technique separates the COVID data into five groups. The Elbow technique results in different records for each cluster, with the first cluster [0] including the most countries (171), the second and third clusters each comprising one country, the fourth cluster containing six countries, and the last cluster containing three countries. It explains why the majority of nations have comparable COVID-affected data and comparable pandemic prevention strategies and measures. The World Health Organization found that when cities and people had easier access to smart city features, the number of cases and deaths in each city reduced, while the rate of recovery and immunization climbed dramatically. Due to the limited data available, the acquired results do not confirm that each smart city responds to pandemics in the same way.