Employing Tweets to Traverse Traffic: GSND and Mobility Patterns

You are here

MTI researchers examine if and how geo-social network data from Twitter can help explain travel patterns
June 10, 2021
|
San José, CA

In 2019, before the pandemic, drivers in the San Francisco Bay urban area lost an average of 47 hours over the year, just sitting in traffic. Sprawling metropolitan regions like the Bay Area deal with traffic nightmares and lengthy commute times on a daily basis. Good transportation planning strives to increase people’s mobility by reducing barriers like cost and time and streamlining travel, but local and regional planners struggle to keep up with rapid changes in mobility patterns. New Mineta Transportation Institute (MTI) research, Comparing Twitter and LODES Data for Detecting Commuter Mobility Patterns, investigates whether geo-social network data (GSND) from geo-tagged tweets can be used to help understand these patterns.

Approximately 33 million geo-referenced Bay Area tweets were harvested for the study period from 2010 until early 2020. They were filtered by repeat occurrences of origin/ destination (O/D) pairs and categorized by time of day and day of week. Each of these pairs, as well as all U.S. Census Bureau Longitudinal Employer-Household Dynamics (LEHD) Origin-Destination Employment Statistics (LODES) O/Ds, were then routed as shortest paths on the Open Street Maps network of roads.

The study found that:

  • While Twitter travel flow patterns occur more regionally—the vast majority of them happen within their census tract and county—over 40% of LODES connections cross county borders. Data analysis suggests that Twitter flows represent shorter trips.
  • GSND can help capture the over 80% of non-commuting trips that keep our roads busy but should be complementary to existing LODES data rather than a substitute.
  • Extremely high correlation rates of street segment use between the two data sources suggests that although LODES data is intended to capture only commuter flows, it is actually an excellent predictor of overall traffic loads for non-rush hour and weekend trips.

“This means that expensive and rarely comprehensive surveys are now only needed to capture trip purposes. Regardless of trip purpose (e.g., shopping, regular recreational activities, dropping kids at school), the LODES data is an excellent predictor of overall road segment loads. The methods we used to capture GSND data are also a step toward a complementary dataset,” explains author Dr. Laxmi Ramasubramanian.

This research suggests further data analysis of geo-tagged tweets to understand trip purposes and a deeper understanding of how mobility patterns have been affected by the pandemic. By conducting sentiment analysis of geo-tagged tweets, researchers could also draw conclusions about why people traveled. This research contributes to understanding when and where people travel and can help planners prepare for shifts in mobility, including quickly changing commuter patterns. Overall, this study is one part of mobility pattern research that helps the transportation professionals and other stakeholders improve mobility for everyone.

This research will be featured in an MTI Research Snap webinar on Thursday, June 17, 2021 at 12p.m. (PT). Guests can register for free at https://tinyurl.com/Traffic617.

 

ABOUT THE MINETA TRANSPORTATION INSTITUTE

At the Mineta Transportation Institute (MTI) at San Jose State University (SJSU) our mission is to increase mobility for all by improving the safety, efficiency, accessibility, and convenience of our nations’ transportation system. Through research, education, workforce development and technology transfer, we help create a connected world. Founded in 1991, MTI is funded through the US Departments of Transportation and Homeland Security, the California Department of Transportation, and public and private grants, including those made available by the Road Repair and Accountability Act of 2017 (SB1). MTI is affiliated with SJSU’s Lucas College and Graduate School of Business.

ABOUT THE AUTHORS
Drs. Laxmi Ramasubramanian and Jochen Albrecht are MTI Research Associates and professors at SJSU and Hunter College, CUNY, respectively. Dr. Bernd Resch is a professor at the University of Salzburg, Austria, as is his PhD student Andreas Petutschnig. Aleisha Wright is a MA student in the SJSU Urban and Regional Planning program.

 

Media Contact:

Irma Garcia

MTI Communications and Operations Manager

O: 408-924-7560

E: Irma.garcia@sjsu.edu

MCTM
CSUTC
NTSC
NTFC

Contact Us

SJSU Research Foundation   210 N. 4th Street, 4th Floor, San Jose, CA 95112    Phone: 408-924-7560   Email: mineta-institute@sjsu.edu