DENTON (UNT), Texas -- Yan Huang, associate professor in the Department of Computer Science and Engineering in the University of North Texas’ College of Engineering, is developing a location-based social media search function that will allow users to search for and find more information about events around them than ever before.
For many social media platforms today, a user can search for an event at a location using keywords, for example, “Concerts in Denton, Texas.” The problem with keyword–based searches is that only social media updates and pages with those exact words will be found.
However, with Huang’s algorithm, a user could search for “Concerts in Denton, Texas,” and find social posts about concerts that originate from Denton, Texas, even when the poster did not mention Denton in their post.
Huang is developing a robust model for detecting events that incorporates check-in data, text data and a user’s hometown location in their social profiles to solve that problem. The model will predict user locations using postings on social platforms and the locations referenced in text on social platforms, and will also rely heavily on how people connect and interact with each other on social networks.
“A tremendous amount of information is being shared every day on social media platforms, including information that is valuable to others such as events, gatherings or even natural disaster occurrences,” Huang said. “With this algorithm, a user will be able to search for events, and results will include tweets or other social updates that may not explicitly mention a location.”
Huang’s research is funded by a Department of Defense grant. She is working with Rada Mihalcea, associate professor in the Department of Electrical Engineering and Computer Science at the University of Michigan.