The GISRUK Seminar Series

Celebrating Innovation in GIS

Beginning in 2021, GISRUK launched a series of international seminars celebrating innovation in Geographical Information Science, Chaired by Peter Mooney. If you would like to hear about up-coming seminars, please sign up to the GISRUK mailing list.

Dr Song Gao, Geospatial Data Science Lab, University of Wisconsin-Madison, USA

Dr Song Gao portrait 3rd November 2021
GeoAI for Human Mobility Analytics and Location Privacy Protection

We have witnessed recent advances in Geospatial Artificial Intelligence (GeoAI), which is the integration of geospatial technologies and AI, especially using machine learning and deep learning methods for geographic knowledge discovery and beyond. The increasing location-based services have generated large-scale individual-level trajectory data through mobile phone tracking, wearable sensors, GPS devices, and social media. Those trajectory big data provide new opportunities to study multiscale human mobility patterns and human-place interactions. It also introduces grand challenges regarding the protection of geoprivacy and broader implications. In this talk, Dr. Gao will present his research group’s latest research efforts on human mobility analytics and protecting user location privacy using various GeoAI approaches (e.g., using recurrent neural networks, generative adversarial networks, and graph convolutional networks).
Download slides here (.pdf)
View Dr Song Gao's Profile
View the Geo Data Science Lab (GeoDSLab@UW-Madison)

Prof. Elena Demidova, The University of Bonn

Dr Song Gao portrait 25th March 2021
Semantic geographic knowledge on a world scale – interlinking OpenStreetMap and knowledge graphs

OpenStreetMap (OSM) is a rich source of openly available volunteered geographic information on a world scale. However, representations of geographic entities in OSM are highly diverse and incomplete. Knowledge graphs (i.e. graph-based knowledge repositories) such as Wikidata, EventKG, and DBpedia are a rich source of contextual semantic information about geographic entities. For example, Wikidata contains over six million geographic entities, including locations, points of interest, mountain peaks, etc. Whereas knowledge graphs provide a wide range of complementary semantic information for geographic entities, interlinking between knowledge graphs and OSM is insufficient with the links mainly manually defined by volunteers. This lecture will introduce emerging approaches that address tighter integration of OSM and knowledge graphs; it gives particular attention to link discovery and semantic enrichment of OSM datasets.
Download slides here (.pdf)
View Prof. Elena Demidova's Profile
View the Data Science & Intelligent Systems (DSIS) research group