Search Results

Documents authored by Wiedemann, Nina

Document
DOI: 10.4230/LIPIcs.GIScience.2023.11
Benchmarking Regression Models Under Spatial Heterogeneity

Authors: Nina Wiedemann, Henry Martin, and René Westerholt

Published in: LIPIcs, Volume 277, 12th International Conference on Geographic Information Science (GIScience 2023)

Abstract
Machine learning methods have recently found much application on spatial data, for example in weather forecasting, traffic prediction, and soil analysis. At the same time, methods from spatial statistics were developed over the past decades to explicitly account for spatial structuring in analytical and inference tasks. In the light of this duality of having both types of methods available, we explore the following question: Under what circumstances are local, spatially-explicit models preferable over machine learning models that do not incorporate spatial structure explicitly in their specification? Local models are typically used to capture spatial non-stationarity. Thus, we study the effect of strength and type of spatial heterogeneity, which may originate from non-stationarity of a process itself or from heterogeneous noise, on the performance of different linear and non-linear, local and global machine learning and regression models. The results suggest that it is necessary to assess the performance of linear local models on an independent hold-out dataset, since models may overfit under certain conditions. We further show that local models are advantageous in settings with small sample size and high degrees of spatial heterogeneity. Our findings allow deriving model selection criteria, which are validated in benchmarking experiments on five well-known spatial datasets.
Cite as

Nina Wiedemann, Henry Martin, and René Westerholt. Benchmarking Regression Models Under Spatial Heterogeneity. In 12th International Conference on Geographic Information Science (GIScience 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 277, pp. 11:1-11:15, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)

Copy BibTex To Clipboard

@InProceedings{wiedemann_et_al:LIPIcs.GIScience.2023.11,
  author =	{Wiedemann, Nina and Martin, Henry and Westerholt, Ren\'{e}},
  title =	{{Benchmarking Regression Models Under Spatial Heterogeneity}},
  booktitle =	{12th International Conference on Geographic Information Science (GIScience 2023)},
  pages =	{11:1--11:15},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-288-4},
  ISSN =	{1868-8969},
  year =	{2023},
  volume =	{277},
  editor =	{Beecham, Roger and Long, Jed A. and Smith, Dianna and Zhao, Qunshan and Wise, Sarah},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GIScience.2023.11},
  URN =		{urn:nbn:de:0030-drops-189064},
  doi =		{10.4230/LIPIcs.GIScience.2023.11},
  annote =	{Keywords: spatial machine learning, spatial non-stationarity, Geographically Weighted Regression, local models, geostatistics}
}
Document
Short Paper
DOI: 10.4230/LIPIcs.GIScience.2023.84
Predicting visit frequencies to new places (Short Paper)

Authors: Nina Wiedemann, Ye Hong, and Martin Raubal

Published in: LIPIcs, Volume 277, 12th International Conference on Geographic Information Science (GIScience 2023)

Abstract
Human mobility exhibits power-law distributed visitation patterns; i.e., a few locations are visited frequently and many locations only once. Current research focuses on the important locations of users or on recommending new places based on collective behaviour, neglecting the existence of scarcely visited locations. However, assessing whether a user will return to a location in the future is highly relevant for personalized location-based services. Therefore, we propose a new problem formulation aimed at predicting the future visit frequency to a new location, focusing on the previous mobility behaviour of a single user. Our preliminary results demonstrate that visit frequency prediction is a difficult task, but sophisticated learning models can detect insightful patterns in the historic mobility indicative of future visit frequency. We believe these models can uncover valuable insights into the spatial factors that drive individual mobility behaviour.
Cite as

Nina Wiedemann, Ye Hong, and Martin Raubal. Predicting visit frequencies to new places (Short Paper). In 12th International Conference on Geographic Information Science (GIScience 2023). Leibniz International Proceedings in Informatics (LIPIcs), Volume 277, pp. 84:1-84:6, Schloss Dagstuhl – Leibniz-Zentrum für Informatik (2023)

Copy BibTex To Clipboard

@InProceedings{wiedemann_et_al:LIPIcs.GIScience.2023.84,
  author =	{Wiedemann, Nina and Hong, Ye and Raubal, Martin},
  title =	{{Predicting visit frequencies to new places}},
  booktitle =	{12th International Conference on Geographic Information Science (GIScience 2023)},
  pages =	{84:1--84:6},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-288-4},
  ISSN =	{1868-8969},
  year =	{2023},
  volume =	{277},
  editor =	{Beecham, Roger and Long, Jed A. and Smith, Dianna and Zhao, Qunshan and Wise, Sarah},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.GIScience.2023.84},
  URN =		{urn:nbn:de:0030-drops-189794},
  doi =		{10.4230/LIPIcs.GIScience.2023.84},
  annote =	{Keywords: Human mobility, Visitation patterns, Place recommendation, Next location prediction}
}
Questions / Remarks / Feedback
X

Feedback for Dagstuhl Publishing


Thanks for your feedback!

Feedback submitted

Could not send message

Please try again later or send an E-mail
© 2023-2024 Schloss Dagstuhl – LZI GmbH About DROPS Imprint Privacy Contact