January 30, 2013, 4:14 PM — Researchers at Johns Hopkins University are using Twitter to track what has been a particularly severe flu season across the U.S.
As people take to Twitter to complain about being sick with the flu and its dreaded symptoms, researchers are using those tweets to gather data about how the influenza epidemic is increasing or ebbing state to state. Since sick people tend to stay home in bed nursing their ailments, Twitter offers a new way to get real-time influenza data.
Chart of the flu pattern across the U.S.Click to view larger image
"This new work demonstrates that Twitter posts can be used to guide public health officials in their response to outbreaks of infectious diseases," Mark Dredze, an assistant research professor in the Department of Computer Science at Johns Hopkins, said in a statement. "Our hope is that the new technology can be used to track other diseases as well."
This has been a particularly bad flu season, according to the Centers for Disease Control and Prevention. Since mid-December, hospitalizations and deaths caused by the flu have "risen steeply," according to the CDC. While the flu has declined in a few parts of the country, it's still high nationally. Between Oct. 1, 2012 and Jan.19, 2013, 6,191 people have been hospitalized because of the flu, the CDC said. Nationwide, 37 children have died from the flu, the agency said.
To get a handle on how the flu is progressing or declining across the country, researchers turned to social networks.
One major problem is that researchers cannot simply count the number of tweets that mention the words "flu" or "influenza" since some people may be asking questions or commenting but may not be sick.
"When you look at Twitter posts, you can see people talking about being afraid of catching the flu or asking friends if they should get a flu shot or mentioning a public figure who seems to be ill," said Dredze, who has been using tweets to monitor public health trends. "But posts like this don't measure how many people have actually contracted the flu. We wanted to separate hype about the flu from messages from people who truly become ill."
To address this problem, Johns Hopkins computer scientists and researchers in the School of Medicine worked together to build a filter with statistical methods based on human language that culls tweets specifically about having the flu.