eprintid: 3365
rev_number: 5
eprint_status: archive
userid: 1
dir: disk0/00/00/33/65
datestamp: 2020-09-17 10:08:43
lastmod: 2020-09-17 10:08:43
status_changed: 2020-09-17 10:08:43
type: thesis
metadata_visibility: show
sword_depositor: 1
creators_name: Kema, Steven
creators_id: S2068222
creators_email: stevenkema@gmail.com
title: Geography and AI: a happy marriage? Exploring the potential of Machine Learning as a new method in geographic research.
ispublished: unpub
full_text_status: public
abstract: Big data analytics can offer Geography new ways of understanding complex socio-spatial processes, especially with the increasingly available amount of data that is produced by society. This thesis explores the potential of machine learning in Geography via a case study of neighbourhood level gentrification prediction. Several machine learning algorithms are compared; XGBoost, CatBoost, and Random Forest regression outperform standard quantitative methods. The implementation of SHapley Additive exPlanations (SHAP) as a way of interpreting machine learning models is explored, and suggests that SHAP is a promising solution to the need for explainable machine learning models. Future gentrification prediction reveals that model specification has substantial impact on result interpretation and practical applicability, and suggests that theoretical foundation remains a key factor in future development of the research field. Machine learning provides a lot of new opportunities for geography but it is also important to be critical of its promises.
date: 2020
pages: 54
thesis_type: master
degree_programme: EG
tutors_name: Koster, S.
tutors_name: Ballas, D.
tutors_organization: Fac. Ruimtelijke wetenschappen, Basiseenheid Economische Geografie
tutors_organization: Fac. Ruimtelijke wetenschappen, Basiseenheid Economische Geografie
tutors_email: Sierdjan.Koster@rug.nl
tutors_email: D.Ballas@rug.nl
security: validuser
keywords_local: Gentrification
keywords_local: Machine Learning
keywords_local: Geographic Data Science
keywords_local: Explainable AI
keywords_local: Big Data
language_iso: en
agreed_repository: yes
date_issued: 2020-09-16
citation:   Kema, Steven  (2020) Geography and AI: a happy marriage? Exploring the potential of Machine Learning as a new method in geographic research.  Master thesis.   
document_url: https://frw.studenttheses.ub.rug.nl/3365/1/thesis-EG-final_Steven-Kema.pdf