Nowcasting COVID-19 hospitalizations using Google Trends and LSTM

Derval, Guillaume;François-Lavet, Vincent;Schaus, Pierre
(2020) AI4SG (20.July.2020)

Files

AI4SG_paper_571.pdf
  • Open Access
  • Adobe PDF
  • 171.44 KB
Download

Details

Authors
  • Derval, GuillaumeUCLouvain
    Author
  • François-Lavet, VincentUCLouvain
    Author
  • Schaus, Pierreorcid-logoUCLouvain
    Author
Abstract
The Google Trends data of some keywords havestrong correlations with COVID-19 hospitaliza-tions. We attempt to use these correlations andshow an experimental procedure using a simpleLSTM model to nowcast hospitalization peaks us-ing Google Trends data. Experiments are done onFrench regions and on Belgium. This is a prelim-inary work, that would need to be tested during a(hopefully non-existing) second peak.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Derval, G., François-Lavet, V., & Schaus, P. (2020). Nowcasting COVID-19 hospitalizations using Google Trends and LSTM. AI4SG 2020. Accepted/in-press. AI4SG. https://hdl.handle.net/2078.5/269947 (Original work published 2020)