Tracking the explosive contagion of Coronavirus sentiment across Networks
We took a brief look at the official data on the spread of the Coronavirus, and ran this against the number of cases per day globally (with around 400 reported cases on 21st January to over 20,000 cases by 3rd February), and those numbers look exponential.
What if we could map the spread of ideas about that virus? As per the great film, Inception: “An idea is like a virus. Resilient. Highly contagious.”
In this case we ran sentiment analysis algorithms tracking media mentions of the virus from 21st January to 3rd February across 22,000 media mentions, tracking sentiment in relation to the virus.
There is a spike in negative media mentions in the United States on 21st January and in Canada on 25th January – the dates of their respective first cases of the disease. Peaks of panic (as seen through the deep red hues) around the globe continue to spike around initial reported cases in various countries.
However the sentiment from those mentions has, surprisingly, become less negative over time: On 31st January, WHO declared the virus a global emergency, the date on which you can see increased concern on a global level (indicated by the orangey red hues that engulf the globe). After that, the tone of the mentions becomes less negative.
This is not to say that the situation itself has improved. On the contrary, the coronavirus is continuing to spread despite many nations putting strict travel bans into place. However the mentions of the virus are now more solution oriented. We are seeing a large increase in keywords such as “Safety” and “Care” – it seems that by identifying an emergency, people are moving to a solutions orientation as people work together to stop the spread.