Communications of the ACM
sing Twitter to Predict the Influence of Lifestyle on Health
University of Rochester researchers are using geographic and other data from Twitter users to model how likely it is that the Twitter users will become sick.
Credit: University of Rochester
University of Rochester researchers have used Twitter to model how factors such as social status, exposure to pollution, and interpersonal interaction can influence health.
The technology enables researchers to passively monitor the population's health trends instead of conducting surveys. Many of the tweets are geo-tagged, which means they carry global positioning system information that shows exactly where the users were when they tweeted, notes Rochester researcher Adam Sadilek.
Using tweets collected in New York City over a period of a month, the researchers examined 70 factors and whether they had a positive, negative, or neutral impact on the users' health. The technology has led to the development of GermTracker, a Web app that color-codes users based on their health by mining information from their tweets for 10 cities around the world.
"This app can be used by people to make personal decisions about their health," Sadilek says. The researchers now are developing a machine-learning algorithm to determine if a tweet indicates the user is sick.
From University of Rochester News
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found