The Natural Language Processing Library is a collection of tools for processing natural language using Python 3. It features tools for manipulating and extracting information from raw English language text, using multiple machine learning and natural language processing algorithms.
Currently it's under active development, and not feature complete. However the core library is nearly feature complete, and will undergo a feature freeze soon. There are plans to put it up on PIP, once the core reaches a reasonable level of stability. There are currently no plans to make this library compatible with older versions of Python.
-
Python 3.3
The library depends on new syntax introduced in Python 3.3, older versions of Python will not work.
-
SQLAlchemy 0.9.1
Currently SQLAlchemy is necessary for the library to work. Though, there are are plans to implement other storage back-ends.
-
Beautiful Soup 4.1.2
This is used for parsing HTML collected by the web-scrapers, everything in the core will run fine without this.
-
NumPy 1.8.0
Everything will run fine without NumPy installed; however, having it installed will make certain neural network algorithms run much faster.
-
NLTK 3.0
There are a few convenience functions that integrate with NLTK. However, it's not entirely necessary.
-
matplotlib 1.2.1
This is only needed to see graphical plots of certain outputs. If it's not installed, the plots simply won't be displayed.
Installation should be straight forward; it installs like pretty much every other Python package.
$ python3 setup.py install