COVID-19 or Flu? Discriminative Knowledge Discovery of COVID-19 Symptoms from Google Trends Data
Google Trends data analytics is gaining more attention in the past few years, and most of the state-of-the-art algorithms are focused on forecasting. How to extract knowledge about symptoms mostly related to COVID-19 by contrasting periods of time with and without the spread of COVID-19 from Google Trends data has not been investigated. To this end, we propose a novel nonnegative discriminative analysis (DNA) to extract the unique information of one dataset relative to another dataset. Numerical tests corroborated the efficacy of our proposed approaches to discover the three unique COVID-19 symptoms w.r.t. flu including ageusia, shortness of breath, and anosmia while prior arts are not able to.
Md Imrul Kaish, Md Jakir Hossain, Evangelos E. Papalexakis, and Jia Chen. 2021. COVID-19 or Flu? Discriminative Knowledge Discovery of COVID-19 Symptoms from Google Trends Data. In epiDAMIK 2021: 4th epiDAMIK ACM SIGKDD International Workshop on Epidemiology meets Data Mining and Knowledge Discovery. ACM, New York, NY, USA, 3 pages.
piDAMIK 2021: 4th epiDAMIK ACM SIGKDD International Workshop on Epidemiology meets Data Mining and Knowledge Discovery