This workshop is designed for beginners eager to develop practical skills in spatial data processing, filtering, and advanced spatial analysis through a real-world case study on nitrate contamination and social vulnerability. Participants will explore how spatial data science can be applied to assess environmental risks and disparities using spatial statistical and machine learning methods. Techniques covered include OLS regression, Spatial Lag Regression, Multiscale Geographically Weighted Regression, and XGBoost with SHAP for explainable machine learning. Participants will also learn to use Google Colab for executing and collaborating on spatial data science workflows, including running Jupyter notebooks in a cloud-based environment and processing large datasets.
Caglar Koylu, associate professor in the Department of Geographical and Sustainability Sciences, will run the workshop.
Register for the workshop here.