Details
| Original language | English |
|---|---|
| Title of host publication | SIGSPATIAL '21 |
| Subtitle of host publication | Proceedings of the 29th International Conference on Advances in Geographic Information Systems |
| Editors | Xiaofeng Meng, Fusheng Wang, Chang-Tien Lu, Yan Huang, Shashi Shekhar, Xing Xie |
| Publisher | Association for Computing Machinery (ACM) |
| Pages | 415-418 |
| Number of pages | 4 |
| ISBN (electronic) | 9781450386647 |
| Publication status | Published - 4 Nov 2021 |
| Event | 29th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, SIGSPATIAL 2021 - Virtual, Online, China Duration: 2 Nov 2021 → 5 Nov 2021 |
Publication series
| Name | GIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems |
|---|
Abstract
OpenStreetMap is a unique source of openly available worldwide map data, increasingly adopted in real-world applications. Vandalism detection in OpenStreetMap is critical and remarkably challenging due to the large scale of the dataset, the sheer number of contributors, various vandalism forms, and the lack of annotated data to train machine learning algorithms. This paper presents Ovid - a novel machine learning method for vandalism detection in OpenStreetMap. Ovid relies on a neural network architecture that adopts a multi-head attention mechanism to effectively summarize information indicating vandalism from OpenStreetMap changesets. To facilitate automated vandalism detection, we introduce a set of original features that capture changeset, user, and edit information. Our evaluation results on real-world vandalism data demonstrate that the proposed Ovid method outperforms the baselines by 4.7 percentage points in accuracy.
Keywords
- Machine Learning, OpenStreetMap, Vandalism Detection
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Earth-Surface Processes
- Computer Science(all)
- Computer Science Applications
- Mathematics(all)
- Modelling and Simulation
- Computer Science(all)
- Computer Graphics and Computer-Aided Design
- Computer Science(all)
- Information Systems
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
SIGSPATIAL '21: Proceedings of the 29th International Conference on Advances in Geographic Information Systems. ed. / Xiaofeng Meng; Fusheng Wang; Chang-Tien Lu; Yan Huang; Shashi Shekhar; Xing Xie. Association for Computing Machinery (ACM), 2021. p. 415-418 (GIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Ovid
T2 - 29th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, SIGSPATIAL 2021
AU - Tempelmeier, Nicolas
AU - Demidova, Elena
N1 - Funding Information: Acknowledgements. This work was partially funded by DFG, The German Research Foundation (“WorldKG”, 424985896), the Federal Ministry for Economic Affairs and Energy (BMWi), Germany (“dE-mand”, 01ME19009B), and the European Commission (EU H2020, “smashHit”, grant-ID 871477).
PY - 2021/11/4
Y1 - 2021/11/4
N2 - OpenStreetMap is a unique source of openly available worldwide map data, increasingly adopted in real-world applications. Vandalism detection in OpenStreetMap is critical and remarkably challenging due to the large scale of the dataset, the sheer number of contributors, various vandalism forms, and the lack of annotated data to train machine learning algorithms. This paper presents Ovid - a novel machine learning method for vandalism detection in OpenStreetMap. Ovid relies on a neural network architecture that adopts a multi-head attention mechanism to effectively summarize information indicating vandalism from OpenStreetMap changesets. To facilitate automated vandalism detection, we introduce a set of original features that capture changeset, user, and edit information. Our evaluation results on real-world vandalism data demonstrate that the proposed Ovid method outperforms the baselines by 4.7 percentage points in accuracy.
AB - OpenStreetMap is a unique source of openly available worldwide map data, increasingly adopted in real-world applications. Vandalism detection in OpenStreetMap is critical and remarkably challenging due to the large scale of the dataset, the sheer number of contributors, various vandalism forms, and the lack of annotated data to train machine learning algorithms. This paper presents Ovid - a novel machine learning method for vandalism detection in OpenStreetMap. Ovid relies on a neural network architecture that adopts a multi-head attention mechanism to effectively summarize information indicating vandalism from OpenStreetMap changesets. To facilitate automated vandalism detection, we introduce a set of original features that capture changeset, user, and edit information. Our evaluation results on real-world vandalism data demonstrate that the proposed Ovid method outperforms the baselines by 4.7 percentage points in accuracy.
KW - Machine Learning
KW - OpenStreetMap
KW - Vandalism Detection
UR - http://www.scopus.com/inward/record.url?scp=85119186288&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2203.11087
DO - 10.48550/arXiv.2203.11087
M3 - Conference contribution
AN - SCOPUS:85119186288
T3 - GIS: Proceedings of the ACM International Symposium on Advances in Geographic Information Systems
SP - 415
EP - 418
BT - SIGSPATIAL '21
A2 - Meng, Xiaofeng
A2 - Wang, Fusheng
A2 - Lu, Chang-Tien
A2 - Huang, Yan
A2 - Shekhar, Shashi
A2 - Xie, Xing
PB - Association for Computing Machinery (ACM)
Y2 - 2 November 2021 through 5 November 2021
ER -