Mark Dredze and Michael Paul, computer scientists at John Hopkins University, collected and assessed 1.5 million tweets, all up to 140 characters in length, posted between May 2009 and October 2010.
Their computer system discarded irrelevant tweets and sorted the data into categories such as flu, allergies, insomnia, cancer and depression.
They then analysed the data to infer how users were treating their illnesses.
READ MORE FROM THE WDM CONTENT NETWORK:
To read the latest edition of Healthcare Global, click here
- Financial crisis triggered increase in Europe suicides
- ‘Pain’ chemical discovery in sunburn could lead to cure
- Britain to launch world’s first IVF lottery game
“We found that some people tweeted that they were taking antibiotics for flu. But antibiotics don't work on the flu, which is a virus, and this practice could contribute to the growing antibiotic resistance problems,” said Paul.
“So these tweets showed us that some serious medical misperceptions exist out there.”
They will present their research at the International Conference on Weblogs and Social Media, which is sponsored by the Association for the Advancement of Artificial Intelligence.
Dredze and Paul have already been in talks with public health scientists, who agree Twitter may be a useful resource for analysing public perceptions of illnesses.
However, as with all social media, Twitter’s use as a research tool is limited.
“In our study we could only learn what people were willing to share,” Paul remarked. “We think there's a limit to what people are willing to share on Twitter.”