RB NLP is an open source (and public domain) project which provides RB developers a range of Natural Language Processing algorithms written in pure RB. NLP algorithms allow researchers and application developers to develop programs which can analyse, parse and even understand natural language, leading to better human computer interfaces, and improvements in current algorithms (such as information retrieval). There is no strict definition or limit on the type of algorithms the project can include, so if you have suggestions or requests (or an offer of help), feel free to contact me using my sourceforge address: firstname.lastname@example.org.
Using the algorithms in your own projects requires no credit or copyright as the project is in the public domain. True freedom doesn't limit anybody from using anything anyway they need.
How do I use it?
At the moment the algorithms implemented are pretty straight forward and self explanatory, so no documentation is currently available. A sample project is included which demonstrates what the algorithms are and how to use them.
To download the latest stable release, click the download link above. Stable releases have been heavily stress tested and passed all unit tests and so should work perfectly (yeah, right), but if they don't, fill out a bug report on the project's sourceforge project page (here).
At the moment no code is stored on the sourceforge CVS servers, though, in the future, this will probably happen. All modules and classes are stored as external XML items which means that CVS should work OK with them. It doesn't really make sense for me to use CVS whilst there's only one developer, however, if you'd like to get involved (and thus hasten the adoption of CVS) send me an email.