Paper: Map Translation Using Geo-tagged Social Media

ACL ID E14-4012
Title Map Translation Using Geo-tagged Social Media
Venue Annual Meeting of The European Chapter of The Association of Computational Linguistics
Session Main Conference
Year 2014

This paper discusses the problem of map translation, of servicing spatial entities in multiple languages. Existing work on entity translation harvests translation ev- idence from text resources, not consider- ing spatial locality in translation. In con- trast, we mine geo-tagged sources for mul- tilingual tags to improve recall, and con- sider spatial properties of tags for transla- tion to improve precision. Our approach empirically improves accuracy from 0.562 to 0.746 using Taiwanese spatial entities.