Crowdsourcing Social Media Data for Emergency Services


Social media is more than just a way to blow off a little steam or to peruse family and acquaintances’ goings-on. It also provides a window into how things work.

Barbara Minsker, civil and environmental engineering chair at SMU Lyle School of Engineering, and her research team are using social media platforms and Big Data to improve the sustainability and resilience of complex environmental and human systems.

One of their research projects uses crowdsourcing data from Waze, the GPS navigation software app owned by Google, to assess street-level flash-flood risk and locate the safest routes for first responders during intense rainfall. The project is funded by a grant from the National Institute of Standards and Technology for the Public Safety Accelerator Innovation Program.

“Right now, the first responder’s routing algorithms assume roads are empty, which is never true in Dallas,” Minsker says. “With the Waze data, we want to provide real-time information to Dallas Fire Rescue by estimating the risk of delay or accident when they send out fire trucks to rescue people.”
Crowdsourcing comes into play when people post street flood alerts on Waze. Minsker’s team identifies where the Waze flood alerts have been reported over a several-year period, compares that data to how much rainfall is recorded and factors in other road characteristics.

“By combining the Waze data with topography and land characteristics, we were able to identify how reliable the Waze flood alerts are and found that 90% are located within 100 feet of depressions that could be prone to flooding,” Minsker adds. The team is now building flood risk models that can then be used for mapping safer routes. Future work with the City of Houston will also incorporate traffic camera videos for further verification of the Waze flood alerts.

Minsker and her students completed another research project that explores how social media data and online stakeholder input can support the design of urban green infrastructure, such as rain gardens. “Social media postings are used to identify where new green infrastructure is located and what people like and don’t like about the installations to help better design green infrastructure spaces,” Minsker explains.

The opportunities for using data like Minsker’s are endless. She and her team have only scratched the surface of how her research can impact social, policy and economic issues in the DFW Region.

This article is part of the 2020 Higher Education Review Magazine.