Correlations between Web Searches and COVID-19 Epidemiological Indicators in Brazil

COVID-19 rapidly spread across the world in an unprecedented outbreak with a massive number of infected and fatalities. The pandemic was heavily discussed searched on internet, which generated big amounts data related to it. This led possibility attempting forecast coronavirus indicators using internet data. For this study, Google Trends statistics for 124 selected search terms were used attempt find keywords had best Spearman correlations lag, as well forecasting model. It found that testing among some others, such “I have contracted covid”, high (≥0.7) few weeks lag (≤4 weeks). Besides that, ARIMAX model those promising results predicting increase or decrease epidemiological indicators, although it not able predict their exact values. Thus, we may be useful outbreaks before they happen, auxiliary tool monitoring disease Brazil.