The repository contains the Winning Submission for the COVID-19 Hackathon at University of Washington, Seattle.
Our Team CuriousDuo were the winners under the category : Storytelling / Data Science
Coronavirus is a family of viruses that causes illness which can vary from common cold and cough which also leads to severe health conditions diarrhea, vomiting, chills, headache and much more. SARS-CoV-2 (n-coronavirus) is the newest entry of the coronavirus family which was discovered in the year 2019. Moreover this virus was not identified in any humans before. Due to the rate of spread, on March 11 2020, the coronavirus (COVID-19) outbreak was declared a global pandemic by WHO. As of now there are over 430k+ deaths and over 3M active cases.
Naturally, twitter users discuss the latest development and react to local developments. However, this pandemic has generated a misinformation pandemic which has influenced the extent of seriousness towards COVID-19 and its policies. This was noticed predominantly in the United States of America, which now has the highest number of COVID-19 cases. While there were states which went into lockdown peacefully and willingly such as Washington, there were other states such as Florida where people vehemently protested against restrictive policies.
We hypothesize that this sentiment will be captured in twitter data and hence can be utilized by the government at all levels to create effective strategies to inform the public better.
For the purpose of this hackathon, we will be focusing on select states for our analysis. The objective is to
- Identify & collect tweets from the states
- Identify the sentiment trends for the state-specific user and how this impacts the spread of COVID-19
Based on the data and our results we can conclude that twitter as an opinion sharing platform accurately captures the sentiment which reflected the public perception towards COVID-19. This influences their attitude towards COVID-19 policies which involve following social distancing and staying at home which if not followed leads to greater spread of infection. Thus, this polar opposite perception of residents of the different states can be caught early to design effective, target informational campaigns to tackle both misinformation and COVID-19 pandemic.