Publications
2020 |
Cutchan, Marvin Mc; Özdal-Oktay, Simge; Giannopoulos, Ioannis Semantic-based urban growth prediction (Journal Article) In: Transactions in GIS, vol. n/a, no. n/a, 2020. (Abstract | Links | BibTeX | Tags: Deep learning, geosemantics, urban growth) @article{doi:10.1111/tgis.12655b, Abstract Urban growth is a spatial process which has a significant impact on the earth’s environment. Research on predicting this complex process makes it therefore especially fruitful for decision-making on a global scale, as it enables the introduction of more sustainable urban development. This article presents a novel method of urban growth prediction. The method utilizes geospatial semantics in order to predict urban growth for a set of random areas in Europe. For this purpose, a feature space representing geospatial configurations was introduced which embeds semantic information. Data in this feature space was then used to perform deep learning, which ultimately enables the prediction of urban growth with high accuracy. The final results reveal that geospatial semantics hold great potential for spatial prediction tasks. |
2019 |
McCutchan, Marvin; Özdal-Oktay, Simge; Giannopoulos, Ioannis Urban Growth Predictions with Deep Learning and Geosemantics (Inproceedings) In: Ehrmann, Katharina; Khosravi, Hamid Reza Mansouri; others, (Ed.): VIENNA Young Scientists Symposium (VSS 2019), pp. 30–31, Book-of-Abstracts.com, Gumpoldskirchen, 2019, ISBN: 978-3-9504017-9-0, (Vortrag: VIENNA Young Scientists Symposium (VSS 2019), Wien; 2019-06-13 -- 2019-06-14). (Abstract | BibTeX | Tags: Deep learning, geosemantics, urban growth) @inproceedings{mccutchan19:30[TUW-286504], This work outlines a novel approach for the prediction of urban growth. The method extracts semantic information of geospatial data and predicts if urban and non-urban areas are going to change in the future, using a deep neural network. The scored prediction accuracy is higher than any other urban growth prediction model. This superiority is based on two novelties: (1) The effective modeling of the geospatial configurations using semantics, (2) the use of deep learning. The proposed method is therefore an effective tool to predict one of the global challenges of urban sprawl and support the future development strategies. |