UCSD - University of California - San Diego

08/04/2026 | Press release | Distributed by Public on 08/04/2026 01:02

UC San Diego Students Learn Spatial Data Science Through Real-World Applications

Published Date

August 04, 2026

Article Content

From public health and environmental monitoring to urban planning and disaster response, organizations are looking to advanced analytics to transform location-based data into actionable insights. To help meet this growing demand, the School of UC San Diego Global Policy and Strategy (GPS) offers GPEC 447: Data Science Approaches to Spatial Analysis, a graduate course taught by Ilya Zaslavsky, director of the San Diego Supercomputer Center (SDSC) Spatial Information Systems Laboratory.

Developed in response to the need for a new generation of spatial data scientists, the course bridges traditional geographic information systems (GIS) training with modern machine learning and data science techniques. The class is the final course in the three-course GPS Spatial Analysis Certificate sequence, which begins with GPEC 443 - GIS and Spatial Data Analysis, then continues with GPEC 444 - Advanced GIS and Remote Sensing and ends with GPEC 447.

"A key motivation was to connect academic learning with real-world applications, allowing students to work with authentic datasets and applied problems in areas such as public health, urban planning, environmental analysis and international development," said Zaslavsky, who also teaches courses through the Halıcıoğlu School of Data Science and Computing (HSDSC) Halıcıoğlu Data Science Institute (HDSI).

Throughout the quarter, UC San Diego students enrolled in Zaslavsky's course work with geospatial datasets and tackle challenges involving international economics, migration, sustainable development, infrastructure conditions, risk forecasting and ecological systems. Through a series of coding projects, they learn how machine learning models can incorporate location-based information to identify patterns, improve predictions and support decision-making.

The course also challenges students to apply these methods to independent research projects. For instance, UC San Diego graduate students Dumas Zhong, Li Shen and Zilu Zhang developed a project examining environmental risks associated with industrial facilities across California. Using Environmental Protection Agency facility records, Toxic Release Inventory (TRI) data, CalEnviroScreen 4.0 metrics and American Community Survey socioeconomic indicators, the team created a facility-level dataset designed to evaluate how spatial context influences environmental risk assessment.

Their analysis found that incorporating spatial relationships improved the identification of facilities that may pose elevated environmental risks to nearby communities.

"For me, the most valuable part of this project was seeing how GIS and machine learning can work together," Zhong said. "Maps help define the problem, spatial features help improve the model and the final predictions can be translated back into geographic priorities."

The course reflects Zaslavsky's broader research focus on developing advanced geospatial information systems and data infrastructure. His work centers on spatial and temporal data integration, large-scale geospatial analytics and scientific cyberinfrastructure. Over his career, he has led the technical development of major initiatives such as the National Hydrologic Information System, helping establish standards that make environmental data more findable, accessible, interoperable and reusable. His research has spanned neuroscience, geology, hydrology and disaster response, where he developed scalable frameworks for integrating diverse and complex spatial datasets.

"As artificial intelligence and data science continue to transform how organizations understand the world around them, courses like DSC 170, a Spatial Data Science upper-division elective I teach at HDSI, and GPEC 447, the graduate course I teach at GPS, help prepare students with the interdisciplinary skills needed to tackle increasingly complex challenges," Zaslavsky said. "By combining machine learning, programming and spatial analysis, students gain practical experience turning geographic data into meaningful insights, which is an increasingly valuable capability in today's data-driven world."

Learn more about research and education at UC San Diego in: Artificial Intelligence,

Read more news about: Halıcıoğlu School of Data Science and Computing, Artificial Intelligence

GIS and Machine Learning Working Together. Credit: Dumas Zhong
UCSD - University of California - San Diego published this content on August 04, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on August 04, 2026 at 07:02 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]