Yep, regex would be the better route. You might find some useful stuff in the code Ben Wing created for the experiments we did on geolocation for Wikipedia and Twitter. Details here:
http://code.google.com/p/textgrounder/wiki/WingBaldridge2011 I'm cc'ing Ben in case he has more direct pointers. Jason On Mon, Jan 30, 2012 at 1:46 PM, Peter Harrington < [email protected]> wrote: > I think you could do that with a regular expression, using the MaxEnt > classifier for this would be a waste of computing resources IMHO. With the > stuff I have done before I can process about 10x the data (for a fixed unit > of time) using regEx over the OpenNLP MaxEnt classifier. > > > On Mon, Jan 30, 2012 at 11:02 AM, Travis Camechis <[email protected] > >wrote: > > > Does anyone know if there are models built for identifying geographical > > coordinates such as Lat/Longs and MGRS? > > > > Thanks, > > Travis > > > -- Jason Baldridge Associate Professor, Department of Linguistics The University of Texas at Austin http://www.jasonbaldridge.com http://twitter.com/jasonbaldridge
