Data-driven mathematical models for assessing the COVID-19: SIRD-type equations

• Ano de defesa: 2021
• Autor(a) principal: Amaral, Fábio Vinícius Goes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Neural networks (computer science); Forecasting; Mass vaccination; Redes neurais (ciência da computação); Previsão; Vacinação em massa
• Link de acesso: http://hdl.handle.net/11449/214330

São Paulo is the most populous state in Brazil, home to around 22% of the country’s population. The total number of COVID-19-infected people in São Paulo has reached more than 1 million, while its total death toll stands at 25% of all the country’s deceases. Join- ing the Brazilian academia efforts in the fight against COVID-19, in this work we describe a unified framework for monitoring and forecasting the COVID-19 progress. A novel fore- casting data-driven method has been proposed, by combining the so-called Susceptible- Infectious-Recovered-Deceased model with machine learning strategies to properly fit the mathematical model’s coefficients for predicting Infections, Recoveries, Deaths, and Viral Reproduction Numbers. We show that the resulting predictor is capable of dealing with badly conditioned data samples while still delivering accurate 10-day predictions. Our in- tegrated computational system can be used for guiding government actions mainly in two basic aspects: real-time data assessment and dynamic predictions of COVID-19 curves for different regions of the state and country. Finally, we extend our methodology to in- vestigate the effects of the vaccination process with a more complex model. In particular, our studies are able to predict different scenarios varying the rate of vaccination and the effectiveness of the vaccines.