INFOMSSML7.5 ECTSQ3EnglishMaster
Spatial statistics and machine learning
FaculteitFaculty of Science
NiveauMaster
Studiejaar2026-2027
Beschrijving
Course goals
- wrangle spatial data (vector and raster), including spatial interpolation
- extract information from remote sensing scenes using different classifications
- analyze spatial data by means of statistical models.
- fit and predict spatial data by geographically weighted local regression.
- find patterns in spatial data using decision trees and random forests algorithms on different spatial dataset.
- adapt machine learning and statistical techniques by analyzing real world problems in the context of group assignments and term project.
- term project report (30% of the final grade)
- project presentation/poster (10%)
- two written exams (30% each)
- assignment (pass or fail)
The final grade is a weighted average of the first three components of the assessment.
To qualify for a repair of the final result the mark needs to be at least a 4, or “AANV”.
Prerequisites
INFOMDWR Data Wrangling.
Prerequisites
Content
In this course, students will learn how to generate meaningful spatial information from various raw data sources.
Students will learn the concepts of spatial statistics e.g. point pattern analysis, spatial interpolation, spatial regression), machine learning with spatial data and visualization.
You will learn to apply these techniques to various spatial data sets including point data, data from sensor networks and remotely sensed data, in computer labs and a final case study (which are executed in groups of three students from different application domains of data science).
These methods are applied to problems in physical geography, human geography, social science, environmental epidemiology.
The course will consist of a series of lectures on specific techniques, followed by a computer practical where the techniques are practiced on a given data set.
The topics addressed in this course include:
- spatial data wrangling of
- vector data: projections, spatial and geometric operations
- raster data: remote sensing (lidar and multispectral data
- spatial interpolation: inverse distance and kriging
- classification of remote sensing images, including nearest neighbor, support vector machines and random forests.
- spatial autocorrelation and clustering
- spatial regression models
- spatial lag model
- spatial error model
- geographically Weighted Regression: neighborhood type, local regression
- machine learning algorithms in spatial modelling
- regression tree
- bagging, random forest, and boosting
- ANN
Classroom lectures, e-lectures, computer labs, assignments, case study project .
During the course, students will start to work on a term project, in which they answer a research question based on their own or a given data set.
The term project team should be carried out by three students.
You will study a real-life problem, which includes the choice of the appropriate technique(s), the appropriate variables, implementing the method, and interpretation of the results.
Literature
- Campbell, J.E., Shin, M. (2012) "Geographic Information System Basics", 248 pp, Open source at lardbucket.org.
- De Jong, S.M., Addink, E.A., Heuff, F. (2015) "Remote sensing lecture notes". 113 pp, Fac. of Geosciences, Utrecht University.
- James, G., Witten, D., Hastie, T., Tibshirani, R. (2013). "An Introduction to Statistical Learning with Applications in R", volume 103, Springer.
- Fischer, M. M., & Getis, A. (Eds.). (2010). "Handbook of applied spatial analysis: software tools, methods and applications" (pp. 599-628). Berlin: springer.
- Lovelace, R., Nowosad, J., & Muenchow, J. (2019). "Geocomputation with R." CRC Press.
- various handout papers.
Reviews0 reviews
Nog geen reviews voor dit vak. Wees de eerste!
Heb jij dit vak gevolgd?
Deel je ervaring met toekomstige studenten. Inloggen met je Universiteit Utrecht mailadres duurt één minuut.
Schrijf een review