An interdisciplinary research team at UB received a grant of $ 378,940 from the US National Science Foundation to explore how to better use the social media platform Twitter to improve disaster response.
Yingjie Hu, assistant professor in the Department of Geography, College of Arts and Sciences, is the principal investigator of the project, with Kenneth Joseph, assistant professor in the Department of Computer and Engineering, School of Engineering and Applied Sciences, as as co-principal investigator.
The team also includes an industry partner, Geocove – a geographic information system (GIS) and disaster management company created by UB alumnus Karyn Senneff Tareen. Graduate and undergraduate students in geography and computer science will be involved.
The frequency of natural disasters and the new use of social media by people to ask for help during these events has revealed the need for more research, according to Hu and Joseph. They cite Hurricane Harvey as the catalyst for the project.
During the emergency, some callers were left on hold for long periods of waiting, according to news reports. The prevalence, familiarity and accessibility of social media make it a powerful tool in helping responders provide assistance effectively, according to Hu and Joseph.
“National Public Radio reported that some people stranded in Houston took to social media for help when they couldn’t reach 911 quickly,” Hu said. “Essentially, we’re looking at how people describe places on social media to ask for help in the context of natural disasters. We will extract the information using AI methods and through a process called ‘geographic analysis’ to translate location mentions from texts into locations on a map. “
Data collected from sample tweets from Hurricane Harvey and other natural disasters will provide insight into how people describe the places. The research results could be of value to emergency responders and cities, and potentially help inform rescue operations in future disasters.
Twitter was selected to review location descriptions because of its accessibility. Hu explains that its open API (application programming interface) allows university researchers to access large datasets. Other social media platforms, such as Facebook and Instagram, are more restrictive, he says.
This whole project involves understanding and extracting descriptions of social media locations during natural disasters using a three-step framework, explains Joseph.
This framework includes understanding descriptions of places during natural disasters; study different geospatial artificial intelligence (GeoAI) approaches for location extraction; and understand spatial biases in place descriptions.
Hu and Joseph will partner with Geocove to study Hurricane Harvey Tweets and categorize them into different categories. The collaboration with Geocove connects Hu and Joseph to emergency managers and municipal data on disaster response.
The research team will develop and integrate AI models to extract and determine the coordinates of the locations where people were asking for help. They will also investigate whether the location descriptions were concentrated in certain areas and determine whether certain neighborhoods were overlooked during the response process.
Although social media users tend to be younger, Hu and Joseph note that younger generations spoke out on behalf of older members of the community to seek help during Hurricane Harvey. Researchers are still working to answer the question “What are we missing?” Also focusing on the demographics of the platform to better determine how to help.
“Such research examines disaster response through the alternative lens of social media. It serves as a basis to inform future response efforts to ultimately reduce inequalities and save lives, ”Hu said.