| Summary: | Since December 2019, the novel coronavirus disease (COVID-19) has had a considerable impact on the health and socio-economic fabric of Italy. The effective reproduction number <i>R<sub>t</sub></i> is one of the most representative indicators of the contagion status as it reports the number of new infections caused by an infected subject in a partially immunized population. The task of predicting <i>R<sub>t</sub></i> values forward in time is challenging and, historically, it has been addressed by exploiting compartmental models or statistical frameworks. The present study proposes an Artificial Neural Networks-based approach to predict the <i>R<sub>t</sub></i> temporal trend at a daily resolution. For each Italian region and autonomous province, 21 daily COVID-19 indicators were exploited for the 7-day ahead prediction of the <i>R<sub>t</sub></i> trend by means of different neural network architectures, i.e., Feed Forward, Mono-Dimensional Convolutional, and Long Short-Term Memory. Focusing on Lombardy, which is one of the most affected regions, the predictions proved to be very accurate, with a minimum Root Mean Squared Error (<i>RMSE</i>) ranging from 0.035 at day <i>t</i> + 1 to 0.106 at day <i>t</i> + 7. Overall, the results show that it is possible to obtain accurate forecasts in Italy at a daily temporal resolution instead of the weekly resolution characterizing the official <i>R<sub>t</sub></i> data.
|
|---|