Tracking Cholera in Haiti: Social Media and Data Mining


Home / Tracking Cholera in Haiti: Social Media and Data Mining

Computers sort data more efficiently: Image by flaivoloka

Social media and the Internet have become a vast source of information for health professionals, evidenced by a study published today in the American Journal of Tropical Medicine and Hygiene that shows how data mining  was used to examine the spread of infectious disease.

Swine Flu to Cholera: Monitoring Disease Online

In 2009, Google monitored the spread of the pandemic H1N1 Swine Flu with the Flu Tracker, for example. Now, by examining social media use and content, a group of researchers from the Children’s Hospital of Boston have found that Internet news sources and social media applications such as Twitter were remarkably effective at tracking the spread of cholera in Haiti after the 2010 earthquake. The study, entitled, Social and news media enable estimation of epidemiological patterns early in the 2010 Haitian cholera outbreak, all started with a website called HealthMap, also funded by Google, which provides real-time monitoring of various diseases worldwide. The research team analyzed data from HealthMap, combined with close to 200,000 Tweets, and used the information to evaluate the spread of the cholera outbreak.

Dr. Rumi Chunara, lead author of cholera-tracking study: Image courtesy of Vincent Auyeung

Interview with Lead Author Rumi Chunara, PhD

Decoded Science had the opportunity to ask Rumi Chunara, of the Informatics Program at Children’s Hospital Boston, Research Fellow at Harvard Medical School, lead author of the study examining this information, a few questions about the project to examine the spread of cholera in Haiti.

Decoded Science: What was the inspiration for this project?

Dr. Chunara: We wanted to see if we could use informal social and news media in a more quantitative way to understand disease dynamics.

Decoded Science:  What was the most challenging aspect of this project?

Dr. Chunara: As with any data type, informal social and news media present their own biases and noise, so we had to figure out how to accommodate for those and best harness the information.

Leave a Comment