Earlier this month, the Geographic Data Service (GeoDS) team took part in the inaugural GeoAI conference in Ghent, Belgium.
The four-day conference brought together researchers and practitioners working at the intersection of geography and artificial intelligence, gathering over attendees from 36 countries and regions.
Kicking off the first day in the session, titled ‘Foundation models and LLMs/VLMs’, was GeoDS Investigator and Lecturer in Urban Data Science and Sustainability at the Edinburgh College of Art, Alessia Calafiore. Alessia’s talk, titled ‘Large Language Models for Time-aware Land Use Classification’, explored the use of LLMs to classify neighbourhoods when fed data about their use at different times of the day, potentially helping to develop novel dynamic classifications of cities that show how their uses change throughout the day.
Mark Green, GeoDS Investigator and Professor of Health Geography at the University of Liverpool, gave a presentation as part of the ‘State of the Art and Outlook of Street-Level Imagery for Urban Science workshop’, where they spoke about our work using Vision Language Models to classify street view imagery to identify food advertising. They also had a poster presentation on ‘Exploring social and spatial inequalities in healthy and sustainable diets using big data: An observational analysis of loyalty card data from a UK Retailer, 2022’.
On the final day of the conference, our Senior Data Scientist, Owen Goodwin, based at the University of Liverpool, presented work with Stef De Sabbata, Associate Professor of Geographical Information Science at the University of Leicester and Smart Data Research UK Fellow, on collaborative work with the Liverpool team (including GeoDS Director Alex Singleton, Professor of Geographic Information Science and Pengyuan Liu, Lecturer in Digital Planning at the University of Glasgow) on building geodemographic classifications with new GeoAI methods that is helping inform how we are building new data products.
The GeoDS team would like to thank the organisers, speakers, and attendees for making the first GeoAI conference such a fantastic event. We look forward to the next one!
