karstenw / linguistics

Several accumulated linguistics modules for Nodebox1 to replace the "en" library.

Linguistics

Several accumulated linguistics modules (pattern, wn, textblob & nltk ) for Nodebox 1 to replace the "en" library.

Additionally there is the new conceptnetreader library which gives access to a local conceptnet database.

These are the installed sources of the libraries. I am currently using Python 3.11.4 but this also worked with 3.8.12

ATTENTION

First Run

Rename the downloaded folder "linguistics" and place it inside the Nodebox "Library" folder.

Before the first run, open and run download_corpora.py inside the linguistics folder. This downloads the needed corpora for nltk, textblob and wn to run. The data folder linguistics-data will be placed in the same folder as the linguistics folder.

nltk corpora for textblob, wn and pattern

The runtime for download_corpora.py is ca. 20 minutes.

If you run it from NodeBox there will be no feedback except the spinning beachball.

Alternatively you can run it from the terminal with a 3.8 or 3.11 python. Perhaps you must install the requirements first.

conceptnet database

For the conceptnet database open and run install-conceptnet-database.py.

The runtime for install-conceptnet-database.py is ca. 10 minutes.

As with the nltk download there is no feedback but it can be run from the terminal to see what's going on.

General

This is a work in progress, a moving target.

Goals:

1. Make it available to Nodebox1

2. Keep it usable from a standard python3

3. Adapt Nodebox1 scripts using "en"

Usage

In a nodebox script do:

import linguistics

After the linguistics import the submodules pattern, wn,nltk & textblob can be accessed with linguistics.nltk or import nltk.