Simple supervised document geolocation with geodesic grids

TLDR

Geolocation can effectively summarize large document collections and is a key component of geographic information retrieval. The study investigates automatic geolocation of documents by identifying their latitude/longitude coordinates using only raw text. The authors employ supervised methods that predict locations on geodesic grids of varying resolution and evaluate them on geotagged Wikipedia articles and Twitter feeds. The best method achieves a median error of 11.8 km on Wikipedia and 479 km on Twitter, improving upon prior results for the Twitter dataset.

Abstract

We investigate automatic geolocation (i.e. identification of the location, expressed as latitude/longitude coordinates) of documents. Geolocation can be an effective means of summarizing large document collections and it is an important component of geographic information retrieval. We describe several simple supervised methods for document geolocation using only the document's raw text as evidence. All of our methods predict locations in the context of geodesic grids of varying degrees of resolution. We evaluate the methods on geotagged Wikipedia articles and Twitter feeds. For Wikipedia, our best method obtains a median prediction error of just 11.8 kilometers. Twitter geolocation is more challenging: we obtain a median error of 479 km, an improvement on previous results for the dataset.