About this course
Course description
This course consolidates essential methods of the data science workflow, introduced in the course Applied Geodata Science I and the book Applied Geodata Science and demonstrates common applications of machine learning methods in Geography and Environmental Sciences by worked examples.
Learning objectives
- Implement a data science workflow for common domain-specific machine learning-based modelling tasks.
- Analyse and visualise models and their predictions and communicate insights.
- Describe the challenges of model fitting and evaluate model generalisability in the context of the prediction task.
- Adopt and benefit from Open Science practices and resources for data-intensive research projects.
Target audience and prerequisites
Master students in Geography and in Climate Sciences at the University of Bern.
Students are expected to master contents taught in 480094 Applied Geodata Science I. If AGDS I has not been attended, it is expected that equivalent knowledge has been gained from comparable courses in other study programs, or that contents of the Applied Geodata Science book are independently studied until October 12, 2023 (before the fourth session in fall semester 2023).
The lead for this course is by by Prof. Benjamin Stocker and co-taught with Dr. Koen Hufkens. The website is a product of the group for Geocomputation and Earth Observation, Institute of Geography, University of Bern. Individual tutorials are co-authored by different contributors.
