Merge pull request nltk#3148 from TiMauzi/wsd/languages

added the option to change the wordnet's language

Author: stevenbird

AUTHORS.md

@@ -296,6 +296,7 @@
 - M.K. Pawelkiewicz <https://github.com/hamiltonianflow>
 - Steven Thomas Smith <https://github.com/essandess>
 - Jan Lennartz <https://github.com/Madnex>
+- Tim Sockel <https://github.com/TiMauzi>
 
 ## Others whose work we've taken and included in NLTK, but who didn't directly contribute it:
 

nltk/wsd.py

@@ -10,14 +10,15 @@
 from nltk.corpus import wordnet
 
 
-def lesk(context_sentence, ambiguous_word, pos=None, synsets=None):
+def lesk(context_sentence, ambiguous_word, pos=None, synsets=None, lang="eng"):
     """Return a synset for an ambiguous word in a context.
 
     :param iter context_sentence: The context sentence where the ambiguous word
          occurs, passed as an iterable of words.
     :param str ambiguous_word: The ambiguous word that requires WSD.
     :param str pos: A specified Part-of-Speech (POS).
     :param iter synsets: Possible synsets of the ambiguous word.
+    :param str lang: WordNet language.
     :return: ``lesk_sense`` The Synset() object with the highest signature overlaps.
 
     This function is an implementation of the original Lesk algorithm (1986) [1].
@@ -36,7 +37,7 @@ def lesk(context_sentence, ambiguous_word, pos=None, synsets=None):
 
     context = set(context_sentence)
     if synsets is None:
-        synsets = wordnet.synsets(ambiguous_word)
+        synsets = wordnet.synsets(ambiguous_word, lang=lang)
 
     if pos:
         synsets = [ss for ss in synsets if str(ss.pos()) == pos]