The system was able to accurately predict coronavirus outbreaks in 14 US states.
According to The Next Web, the artificial intelligence created by New York-based risk identification firm Dataminr was able to predict where the next surge in UK and US coronavirus cases would be by analyzing social media posts.
According to the company's website, “the growing number of on-site eyewitness clusters, first-hand on social media on COVID-19,” enabled their algorithm to detect hotspots 7-15 days before COVID-19 exponential growth.
These social media posts include 'posts from people who indicated they tested positive, people showing they are experiencing symptoms' and so on.
AI also predicted future surges in 14 different US states. Seven days later, the coronavirus has hit all 14 states, according to TNW.
“Posts like these seem to be a major indicator of where cases may rise in the future,” Dataminr CEO Ted Bailey told Business Insider.
Analyzing social media to predict outbreaks can better prepare for a crisis.
But any AI predictions must be taken with caution. No artificial intelligence in the world can confirm with 100 percent accuracy the validity of claims by social media users about positive test results or symptoms in relatives.
Sources: Photo: MS TECH / UNSPLASH
