Participation in a pre-course on Python etc (April 3-4) is suggested. More information can be found in the moodle course "Machine learning".
The aim of this module is to gain theoretical and practical knowledge on concepts and methods of spatial modeling of the human-environment system using machine learning techniques.The course will comprise introductory lectures, hand's on exercises,discussions and a large share of project group work for a group competition.
This year, transdisciplinary groups of Geography and Computer Science students will be formed. You will use multiple different datasets (e.g. remote-sensing, statistical, census etc.) from case studies and apply exploratory spatial data analysis and data-driven machine-learning techniques (MXNet Framework, Gluon). Results of the group competition work will be presented and discussed in the last session (including a price for the best team).
The final report (may be written by single or more authors) is due to end of September.
Prior knowledge in statistics and/or spatial analysis and/or programming is expected. |
