A deep-learning model for evaluating and predicting the impact of lockdown policies on COVID-19 cases

TL;DR

This paper introduces a deep-learning model that evaluates and predicts the effects of different lockdown policies on COVID-19 case numbers by clustering countries and analyzing policy impacts.

Contribution

It presents a novel approach combining clustering and deep learning to assess and forecast COVID-19 cases under various lockdown scenarios.

Findings

Lifting restrictions on schools and borders increases COVID-19 cases significantly.

The model achieves competitive prediction accuracy on Qatar data.

Scenario analysis helps evaluate policy impacts effectively.

Abstract

To reduce the impact of COVID-19 pandemic most countries have implemented several counter-measures to control the virus spread including school and border closing, shutting down public transport and workplace and restrictions on gathering. In this research work, we propose a deep-learning prediction model for evaluating and predicting the impact of various lockdown policies on daily COVID-19 cases. This is achieved by first clustering countries having similar lockdown policies, then training a prediction model based on the daily cases of the countries in each cluster along with the data describing their lockdown policies. Once the model is trained, it can used to evaluate several scenarios associated to lockdown policies and investigate their impact on the predicted COVID cases. Our evaluation experiments, conducted on Qatar as a use case, shows that the proposed approach achieved competitive prediction accuracy. Additionally, our findings highlighted that lifting restrictions particularly on schools and border opening would result in significant increase in the number of cases during the study period.