Working on my Java NLP library that can figure out what geographical location is from a piece of free text.

e.g.

"san francisco ca united states" - San Francisco, California, America

"california, san francisco" - San Francisco, California, America

"glasgow, kentucky" - Glasgow, Kentucky, America

"glasgow, UK" - Glasgow, Scotland, United Kingdom

It started as a project when I was scraping websites, and some data had inconsistent ways of writing a location or address.

The library is called location4j - https://github.com/tomaytotomato/location4j