资 源 简 介
Context
This project aims at studying the context of two words in French : "Ton" and "Voix" ("voice" and "tone/pitch"). The question is : what are the most plausible adjectives with those names ? To answer this question I made statistics evaluating which kind of adjectives were associated to those words. I had 100Mo of novels extracted from gutenberg project.
How
For extracting the adjectives associated to "voix" and "ton" I used a parser : the xerox parser online :
http://open.xerox.com/Services/XIPParser/Pages/Using%20XIP.
I made a script that send the corpora and collect the information needed :
Name Adjective1 Occurences Adjective2 Occurences
Example :
Voix : aigüe 33 grave 42 ...
Why did-I use Xerox parser ?
For knowing which adjective is associated to a name, we need a deep parsing. Xerox is a