Next Article in Journal
Combining Global Geopotential Models, Digital Elevation Models, and GNSS/Leveling for Precise Local Geoid Determination in Some Mexico Urban Areas: Case Study
Previous Article in Journal
PSOTSC: A Global-Oriented Trajectory Segmentation and Compression Algorithm Based on Swarm Intelligence
Previous Article in Special Issue
An Innovative Intelligent System with Integrated CNN and SVM: Considering Various Crops through Hyperspectral Image Data
Article

Deep Learning for Toponym Resolution: Geocoding Based on Pairs of Toponyms

1
INSA Lyon, LIRIS UMR CNRS 5205, 69100 Villeurbanne, France
2
Instituto Superior Técnico and INESC-ID, University of Lisbon, 1049-001 Lisbon, Portugal
*
Author to whom correspondence should be addressed.
Academic Editors: Wolfgang Kainz, Davide Buscaldi and Eric Kergosien
ISPRS Int. J. Geo-Inf. 2021, 10(12), 818; https://doi.org/10.3390/ijgi10120818
Received: 30 September 2021 / Revised: 16 November 2021 / Accepted: 28 November 2021 / Published: 2 December 2021
Geocoding aims to assign unambiguous locations (i.e., geographic coordinates) to place names (i.e., toponyms) referenced within documents (e.g., within spreadsheet tables or textual paragraphs). This task comes with multiple challenges, such as dealing with referent ambiguity (multiple places with a same name) or reference database completeness. In this work, we propose a geocoding approach based on modeling pairs of toponyms, which returns latitude-longitude coordinates. One of the input toponyms will be geocoded, and the second one is used as context to reduce ambiguities. The proposed approach is based on a deep neural network that uses Long Short-Term Memory (LSTM) units to produce representations from sequences of character n-grams. To train our model, we use toponym co-occurrences collected from different contexts, namely textual (i.e., co-occurrences of toponyms in Wikipedia articles) and geographical (i.e., inclusion and proximity of places based on Geonames data). Experiments based on multiple geographical areas of interest—France, United States, Great-Britain, Nigeria, Argentina and Japan—were conducted. Results show that models trained with co-occurrence data obtained a higher geocoding accuracy, and that proximity relations in combination with co-occurrences can help to obtain a slightly higher accuracy in geographical areas with fewer places in the data sources. View Full-Text
Keywords: toponym resolution; geocoding; deep neural networks toponym resolution; geocoding; deep neural networks
Show Figures

Figure 1

MDPI and ACS Style

Fize, J.; Moncla, L.; Martins, B. Deep Learning for Toponym Resolution: Geocoding Based on Pairs of Toponyms. ISPRS Int. J. Geo-Inf. 2021, 10, 818. https://doi.org/10.3390/ijgi10120818

AMA Style

Fize J, Moncla L, Martins B. Deep Learning for Toponym Resolution: Geocoding Based on Pairs of Toponyms. ISPRS International Journal of Geo-Information. 2021; 10(12):818. https://doi.org/10.3390/ijgi10120818

Chicago/Turabian Style

Fize, Jacques, Ludovic Moncla, and Bruno Martins. 2021. "Deep Learning for Toponym Resolution: Geocoding Based on Pairs of Toponyms" ISPRS International Journal of Geo-Information 10, no. 12: 818. https://doi.org/10.3390/ijgi10120818

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop