Data driven time-varying SEIR-LSTM/GRU algorithms to track the spread of COVID-19

Abstract

COVID-19 is an infectious disease caused by a newly discovered coronavirus, which has become a worldwide pandemic greatly impacting our daily life and work. A large number of mathematical models, including the susceptible-exposed-infected-removed (SEIR) model and deep learning methods, such as long-short-term-memory (LSTM) and gated recurrent units (GRU)-based methods, have been employed for the analysis and prediction of the COVID-19 outbreak. This paper describes a SEIR-LSTM/GRU algorithm with time-varying parameters that can predict the number of active cases and removed cases in the US. Time-varying reproductive numbers that can illustrate the progress of the epidemic are also produced via this process. The investigation is based on the active cases and total cases data for the USA, as collected from the website "Worldometer". The root mean square error, mean absolute percentage error and $ r_2 $ score were utilized to assess the model's accuracy.

Keywords

• seir
• lstm
• gru
• time-varying parameters
• data-driven
• covid-19
• time-varying reproduction number