Researchers at the University of Missouri have created an artificial intelligence (AI) tool to analyze public Instagram posts to improve city safety, services, and emergency response efforts. The study was published in Frontiers in Computer Science.
An example of mapping urban sentiment hotspots as visualized through an AI-drive framework. Image Credit: Jayedi Aman
Jayedi Aman, an Assistant Professor of Architectural Studies at the University of Missouri, views cities through human movement and connection with urban spaces, suggesting that human experience and physical materials could significantly influence future city design.
In a recent study, Aman and Tim Matisziw, a Professor of Geography and Engineering at the University of Missouri, pioneered a novel approach to urban research by employing artificial intelligence to investigate the emotional dimension of city life. Their objective was to better understand the relationship between a city's physical characteristics and the emotions people experience within those settings.
Analyzing public Instagram posts with location tags, the researchers trained an AI tool to interpret the emotional sentiment expressed in the images and text, categorizing feelings as happy, frustrated, or relaxed. Subsequently, they analyzed the visual attributes of these locations in the real world using Google Street View and a second AI tool. They correlated these features with people's emotions during their social media posts.
This analysis enabled Aman and Matisziw to develop a digital "sentiment map" illustrating the emotional landscape of a city. Their next step involves utilizing this information to create a digital replica of a city – an urban digital twin – capable of displaying people's real-time emotional states.
This form of emotional mapping provides city leaders with a potent new instrument. Rather than depending solely on time-consuming surveys, which may not capture the sentiments of the entire population, this AI-driven approach utilizes data that people are already publicly sharing online.
For example, if a new park gets lots of happy posts, we can start to understand why. It might be the green space, the quiet nature or the sense of community. We can now connect those feelings to what people are seeing and experiencing in these places.
Jayedi Aman, Study Leader, University of Missouri
Extending beyond parks, this tool has the potential to assist officials in enhancing public services, pinpointing areas where individuals feel insecure, planning for emergencies, and monitoring community well-being following disasters.
AI doesn’t replace human input. But it gives us another way to spot patterns and trends that we might otherwise miss, and that can lead to smarter decisions.
Tim Matisziw, Professor, Geography and Engineering, University of Missouri
The researchers envision that this information regarding public sentiment could eventually be displayed alongside traffic and weather updates on digital platforms used by city officials to inform their operational decisions.
“We envision a future where data on how people feel becomes a core part of city dashboards. This opens the door to designing cities that not only work well but also feel right to the people who live in them,” said Aman.
Journal Reference:
Aman, J., et al. (2025) Urban sentiment mapping using language and vision models in spatial analysis. Frontiers in Computer Science. doi.org/10.3389/fcomp.2025.1504523