Student information
MSc thesis topic: Machine learning for ecosystem service models
Ecosystem service models support these policies, but they are generally poorly validated, especially at large scales, which undermines their credibility (Willcock et al 2019). Ecosystem service models include several sources of uncertainty.
A machine learning approach will be developed and applied for learning from ecosystem service model responses. The approach will be evaluated in a case study in sub-Saharan Africa.
Objectives
- Prepare/process spatial ecosystem service data for machine learning
- Develop a machine learning pipeline
- Evaluate the machine learning approach in a case study
Literature
- Willcock, S., et al. (2019). A Continental-Scale Validation of Ecosystem Service Models. Ecosystems , 22(8), 1902–1917.
- Willcock, S.,et al. (2018). Machine learning for ecosystem services. Ecosystems , 33, 165–174.
Requirements
- FTE-35306 Machine Learning
- GRS-34806 Deep Learning
Theme(s): Modelling & visualisation