Open source natural language processing?

1 pointlimeade2 comments

I'm new to natural language processing and was wondering if anyone knows of a good place to start.

What specifically are you trying to do and what language are you using?

For Python, I like NLTK at http://nltk.sourceforge.net

Java has something similar, though I've not used it, at http://opennlp.sourceforge.net/

There are tons of other Java libs around, per usual.

I know of a few Lisp and C/C++ libs, but I've never used Ruby or PHP, etc.

As far as reference books, I mainly just use online documentation and references (Google it.) I also like an older Lisp book by Mellish, but it doesn't use the latest techniques or anything.

Thanks for your response! I'm trying to: (1) parse questions into the "opening" such as "What is the", and "subject" like "capital of Burkina Faso"? (2) also looking for similarities between things like "going running" and "going for a jog" I'm not sure if both of these fall under natural processing... mostly I use python so your first suggestion sounds like a good starting place.

Reply on news.ycombinator.com