News Sentiment Informed Time-series Analyzing AI (SITALA) to curb the spread of COVID-19 in Houston

作者：

Highlights：

• First successful implementation of multivariate CNN to forecast COVID-19 spread.

• The CNN model accepts COVID-19 test positivity and news sentiment as inputs.

• COVID-19 news sentiment is obtained using IBM’s Watson Discovery News.

• The county-level model can aid public policymakers to curb the spread of COVID-19.

• The model predictions fare better than a published Bayesian-based SEIRD model.

摘要

•First successful implementation of multivariate CNN to forecast COVID-19 spread.•The CNN model accepts COVID-19 test positivity and news sentiment as inputs.•COVID-19 news sentiment is obtained using IBM’s Watson Discovery News.•The county-level model can aid public policymakers to curb the spread of COVID-19.•The model predictions fare better than a published Bayesian-based SEIRD model.

论文关键词：COVID-19 model,News sentiment,Public policy,Deep learning,Artificial intelligence,Pandemic forecast

论文评审过程：Received 25 November 2020, Revised 21 April 2021, Accepted 21 April 2021, Available online 29 April 2021, Version of Record 6 May 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115104