Where are they going? Using social media to understand human movement and disease transmission

Share
Date: 
Monday, August 3, 2015 - 2:57pm

 

movement patterns captured in tweets

Movement patterns captured at different temporal scales illustrate connectivity between

districts in Kenya within a 24-hour time period (N = 90,645 tracks) and during a 10-month time period (N = 17,900). Each line represents a movement segment within the country.

 

The emergence of new diseases and the re-emergence of old diseases, leading causes of death worldwide, are an increasing challenge. Recent examples include Ebola in Sierra Leone and Guinea, Polio in Somalia and Cameroon, and MERS in the Arabian Peninsula, according to the Centers for Disease Control.

 

Disease does not walk down the road by itself; it is carried by humans. An integral part of defining how diseases are spread comes from understanding human movement in places where it is difficult to track.

 

The findings from a recent study conducted by Justine I. Blanford, Zhuojie Huang, Alexander Savelyev, and Alan M. MacEachren, published on PLOS ONE, show that  Twitter can be used to capture human mobility at a level of detail  and timeliness not previously possible using other means such as cell phones, dollar bill tracking, or airline traffic data.

 

Read the research article here:

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0129202

 

“One way of trying to capture human movement has been through the use of cell phone data. But, these data are difficult and costly to get, so I started looking at alternative sources. We have been using Twitter data for visualizing emergency crisis situations here in the GeoVISTA center. So we started by looking at an area where cell phone data has been used to capture human movement and that led us to Kenya, where 62 percent of the population has access to a mobile phone,” Blanford explained.

 

There are other drawbacks to using cell phone calling data, Blanford explained. The phone calls are captured to the nearest cell phone tower location and are generally restricted to country-level, making cross-border human movement difficult to capture. Many social media apps, like Twitter, have geo-location features that users can enable to attach location information in the form of coordinates or place names. When individuals tweet, they leave a digital “geographic footprint” that has a more precise GPS-generated location rather than the location of the nearest cell tower and can be freely collected.

 

For this 10-month-long study, 720,149 tweets were captured in Kenya resulting in an average of 4,931 tweets collected per day. Geo-referenced tweets arranged in time by each person can show semi-continuous movement for that individual and, due to the volume, can provide key insights into human movement patterns. When movement was analyzed on a daily basis, Blanford and her colleagues found that most people traveled short distances (less than five kilometers or three miles) and within small areas. Over time, however, they found that people traveled further away, sometimes crossing national boundaries.

 

The study also investigated the ability to map cross-border movement by analyzing an additional 417,451 tweets.  Approximately, three percent of the users who tweeted from Kenya also tweeted from the surrounding countries such as Uganda, Tanzania, Somalia and South Sudan. Analysis of the movements recorded via Twitter found that eighty-five percent of these users only traveled between two countries, eleven percent between three countries, two percent between four countries and only two people visited all five countries.

 

“This research is important for not only examining cross-border movement and importation of disease but also for identifying disease epicenters by determining regional travel hubs that people pass through,” Blanford said. “For example, in light of the current Ebola outbreak, identifying major transport hubs is important for determining places that may be at high risk for the transmission of the virus.”

 

This work also has implications beyond epidemiology. At the end of 2014, 59.5 million people —the highest number on record—were forcibly displaced around the globe due to conflict, persecution, population growth, urbanization, food and energy insecurity, water scarcity, and climate change, according to the United Nations High Commission for Refugees. These people may have been displaced or fled their home country.  But in order to help refugees, agencies need to know where they are.