Multi-Level Gazetteer-Free Geocoding

Sayali Kulkarni, Shailee Jain, Mohammad Javad Hosseini, Jason Baldridge, Eugene Ie, Li Zhang


Abstract
We present a multi-level geocoding model (MLG) that learns to associate texts to geographic coordinates. The Earth’s surface is represented using space-filling curves that decompose the sphere into a hierarchical grid. MLG balances classification granularity and accuracy by combining losses across multiple levels and jointly predicting cells at different levels simultaneously. It obtains large gains without any gazetteer metadata, demonstrating that it can effectively learn the connection between text spans and coordinates—and thus makes it a gazetteer-free geocoder. Furthermore, MLG obtains state-of-the-art results for toponym resolution on three English datasets without any dataset-specific tuning.
Anthology ID:
2021.splurobonlp-1.9
Volume:
Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics
Month:
August
Year:
2021
Address:
Online
Editors:
Malihe Alikhani, Valts Blukis, Parisa Kordjamshidi, Aishwarya Padmakumar, Hao Tan
Venue:
splurobonlp
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
79–88
Language:
URL:
https://aclanthology.org/2021.splurobonlp-1.9
DOI:
10.18653/v1/2021.splurobonlp-1.9
Bibkey:
Cite (ACL):
Sayali Kulkarni, Shailee Jain, Mohammad Javad Hosseini, Jason Baldridge, Eugene Ie, and Li Zhang. 2021. Multi-Level Gazetteer-Free Geocoding. In Proceedings of Second International Combined Workshop on Spatial Language Understanding and Grounded Communication for Robotics, pages 79–88, Online. Association for Computational Linguistics.
Cite (Informal):
Multi-Level Gazetteer-Free Geocoding (Kulkarni et al., splurobonlp 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.splurobonlp-1.9.pdf