Source code for the article "Vers une estimation robuste des proportions lexicales" (2020)

The script file "text_lexical_proportions_analysis.py" contains the main code use in the article "Vers une estimation robuste des proportions lexicales" (submitted to JADT 2020).

This code will analyse a text file (by default, the Jule Vernes novel "De la Terre à la Lune" found on http://www.gutenberg.org/files/799/799-0.txt) to compute lexical proportions for different word properties. It will need a file containing french stopwords (we use here the file provided by Jacques Savoy http://members.unine.ch/jacques.savoy/clef/frenchST.txt).

The 3 lexical proportions studied in this code are :

1. The proportion of stopwords.
2. The proportion of words longer than the mean length.
3. The proportion of words with an even length.

Link to the article will come in the future.