A place name in a text may have more than one potential referent (e.g. Peru the country vs. Peru the city in Indiana). The Edinburgh Geoparser is a system to automatically recognise place names in text and disambiguate them with respect to a gazetteer. The geoparser can be used with several gazetteers including Unlock and GeoNames. It can also be used to process a variety of input text processes. The demonstration version displays output with a Google maps visualisation. Edina's Unlock text/places service provides another demonstration version. An open source release is planned for 2011.

The Edinburgh Geoparser has been developed as part by a number of projects and applied to a range of data sets:

• GeoCrossWalk
• BOPCRIS
• GeoDigRef (Histpop, BOPCRIS, BL)
• Embedding GeoCrossWalk (Stormont Papers)
• SYNC3 (online news)
• Chalice (EPNS)

Publications:

• Claire Grover, Richard Tobin, Kate Byrne, Matthew Woollard, James Reid, Stuart Dunn, and Julian Ball. 2010b. Use of the Edinburgh Geoparser for georeferencing digitised historical collections. Philosophical Transactions of the Royal Society A.
• Richard Tobin, Claire Grover, Kate Byrne, James Reid, and Jo Walsh. 2010. Evaluation of georeferencing. In Proceedings of the 6th Workshop on Geographic Information Retrieval (GIR'10), Zurich, Switzerland.
• Bea Alex and Claire Grover. 2010. Labelling and spatio-temporal grounding of news events. In Proceedings of the workshop on Computational Linguistics in a World of Social Media at NAACL 2010, Los Angeles, USA.
• Presentation at Pelagious workshop, March 2011.