combine_words_through_grammar replaced with stem_words function using nltk stemmer

app/difficulty.py

@@ -7,6 +7,7 @@
 
 import pickle
 import math
+from nltk.stem import WordNetLemmatizer
 from wordfreqCMD import remove_punctuation, freq, sort_in_descending_order, sort_in_ascending_order
 
 
@@ -75,27 +76,11 @@ def revert_dict(d):
     return d2
 
 
-def combine_words_through_grammar(lst,d): #通过语法合并同一单词的不同形式
+def stem_words(list_of_words):  # It reduces words to the root word (eg. ate, eaten -> eat; leaves, leaf -> leaf)
-    lst1=lst
+    wnl = WordNetLemmatizer()
-    for index,word in enumerate(lst):
+    lst1 = [wnl.lemmatize(w) for w in list_of_words]
-        change_word=''
+    return [wnl.lemmatize(w, pos='v') for w in lst1]  # stem by verb: 'v' represents verb
-        if word.endswith('ies'):      #语法条件匹配
+
-            change_word=word[:-3]+'y'
-        elif word.endswith('es'):
-            change_word=word[:-2]
-        elif word.endswith('s'):
-            change_word=word[:-1]
-        elif word.endswith('ed'):
-            change_word=word[:-2]
-        elif word.endswith('en'):
-            change_word=word[:-2] + 'an'
-        else:
-            pass
-        for word2 in d:
-            if change_word==word2:
-                lst1[index]=change_word
-                break
-    return lst1   
 
 def user_difficulty_level(d_user, d):
     d_user2 = revert_dict(d_user) # key is date, and value is a list of words added in that date
@@ -104,7 +89,7 @@ def user_difficulty_level(d_user, d):
     for date in sorted(d_user2.keys(), reverse=True): # most recently added words are more important while determining user's level
         lst = d_user2[date] # a list of words
         #print(lst)
-        lst=combine_words_through_grammar(lst,d) #合并单词的不同形式
+        lst = stem_words(lst) # this call returns a list of words reduced to root word
         #print(lst)
         lst2 = [] # a list of tuples, (word, difficulty level)
         for  word in lst:
@@ -116,7 +101,7 @@ def user_difficulty_level(d_user, d):
         for t in lst3:
             word = t[0]
             hard = t[1]
-            #print('WORD %s HARD %4.2f' % (word, hard))
+            print('WORD %s HARD %4.2f' % (word, hard))
             geometric = geometric * (hard)
             count += 1
             if count >= 10: