Release Date: January 26, 2017 This content is archived.
BUFFALO, N.Y. — After a long wait, the train arrives. But it’s stuffed with baseball fans headed to the game. There is no room to board. The doors slide shut and the wait continues.
It’s a situation that frustrates even the most hardened subway riders.
In a preliminary study, University at Buffalo engineers found that as subway use swells during events that draw big crowds, so too does the number of tweets at these events. The results suggest that data from Twitter, and possibly other social media platforms, can be used to improve event planning, route scheduling, crowd regulations and other subway operations.
“Social media offers a cost-effective way to obtain real-time data on monitoring subway passenger flow,” says Qing He, PhD, Stephen Still Assistant Professor in Transportation Engineering and Logistics at UB, and the study’s corresponding author. “Our results show that data from apps like Twitter can help public transportation officials prepare for and react to passenger surges during concerts, baseball games and other big events.”
In addition to He, who has appointments in UB’s Department of Civil, Structural and Environmental Engineering and the Department of Industrial and Systems Engineering, co-authors are Jing Gao, PhD, assistant professor in UB’s Department Computer Science and Engineering, and Ming Ni, PhD candidate at UB Department of Industrial and Systems Engineering. All are researchers in UB's School of Engineering and Applied Sciences.
To conduct the study, the researchers gathered subway ridership information from April to October in 2014 via turnstiles at Mets-Willets Point station in Queens, New York. They chose the station because it’s located next to Citi Field, the home of Major League Baseball’s New York Mets, and the USTA Billie Jean King National Tennis Center, where the U.S. Open tennis championships are held each year.
The researchers also collected nearly 30 million tweets geotagged to the New York City area during the same time. They then filtered the tweets by their geographic coordinates (a feature that Twitter users enable on their accounts), the context of the tweet (for example, #subwayseries or #USOPEN), the time and other pertinent elements.
Using six different computer models, the researchers then analyzed the data and found what they describe as a moderate positive correlation between passenger flow and the rates of tweets during big events.
“The results are encouraging for two reasons. First, they indicate that increases in social media posts and subway ridership can be linked. Secondly, we have developed a method to track this correlation,” says Gao. “Now, the challenge is to refine this method so it can be used by public transit system operators to improve their systems.”
An early version of the study, “Forecasting the Subway Passenger Flow under Event Occurences with Social Media,” was published online in October in the journal IEEE Transactions on Intelligent Transportation Systems. The study will appear in an upcoming print edition of the journal.
Director of News Content
Engineering, Computer Science