Abstract
Geolocation is the task of identifying a social media user’s primary location, and in natural language processing, there is a growing literature on to what extent automated analysis of social media posts can help. However, not all content features are equally revealing of a user’s location. In this paper, we evaluate nine name entity (NE) types. Using various metrics, we find that GEO-LOC, FACILITY and SPORT-TEAM are more informative for geolocation than other NE types. Using these types, we improve geolocation accuracy and reduce distance error over various famous text-based methods.- Anthology ID:
- W17-4415
- Volume:
- Proceedings of the 3rd Workshop on Noisy User-generated Text
- Month:
- September
- Year:
- 2017
- Address:
- Copenhagen, Denmark
- Venue:
- WNUT
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 116–121
- Language:
- URL:
- https://aclanthology.org/W17-4415
- DOI:
- 10.18653/v1/W17-4415
- Cite (ACL):
- Bahar Salehi, Dirk Hovy, Eduard Hovy, and Anders Søgaard. 2017. Huntsville, hospitals, and hockey teams: Names can reveal your location. In Proceedings of the 3rd Workshop on Noisy User-generated Text, pages 116–121, Copenhagen, Denmark. Association for Computational Linguistics.
- Cite (Informal):
- Huntsville, hospitals, and hockey teams: Names can reveal your location (Salehi et al., WNUT 2017)
- PDF:
- https://preview.aclanthology.org/starsem-semeval-split/W17-4415.pdf