“This aggregated, anonymized dataset shows trends in search patterns for symptoms and is intended to help researchers to better understand the impact of COVID-19. Public health experts indicated that trends in search patterns might be helpful in broadly understanding how COVID-19 impacts communities and even in detecting outbreaks earlier. You shouldn’t assume that the data is a recording of real-world clinical events, or use this data for medical diagnostic, prognostic, or treatment purposes. This data reflects the volume of Google searches for a broad set of symptoms, signs and health conditions. To keep things simple in this documentation, we will refer to all of these collectively as symptoms. The data covers hundreds of symptoms such as fever, difficulty breathing, and stress—based on the following:

• a symptom’s prevalence in Google’s searches
• data quality and privacy considerations

For each day, we count the searches mapped to each of these symptoms and organize the data by geographic region. The resulting dataset is a daily or weekly time series for each region showing the relative frequency of searches for each symptom…”