自定义词形修饰词并附加到WordNetLemmatizer

Question

我想为定形结果附加一些例外。例如，当我测试wnl.lemmatize('cookies')时，我得到的结果是cooky而不是cookie。如何将去词化结果更新为cookie？

import nltk
from nltk.tokenize import word_tokenize
from nltk import pos_tag
from nltk.stem import WordNetLemmatizer 
wnl = WordNetLemmatizer()

def text_cleaning(text):
  text = text.lower()
  tok_list = [wnl.lemmatize(w,tag[0].lower()) if tag[0].lower() in ['a','n','v'] else wnl.lemmatize(w) for w,tag in pos_tag(word_tokenize(text))]
return ' '.join(tok_list)

Answer 1

仔细查看here的实现，您可能可以做类似

的操作

class WNWrapper(WordNetLemmatizer):
    def __init__(self, custom_transforms):
        self.custom_transforms = custom_transforms

    def lemmatize(self, word):
        if word in self.custom_transforms:
            return self.custom_transforms[word]
        super().lemmatize(word)

但这仅在

时有效

1）您知道要更改/不更改哪些词

2）这是一个很小的数字。这显然无法扩展