除非 POS 显式，否则 WordNetLemmatizer 不会返回正确的引理 - Python NLTK

我正在对 Ted 数据集成绩单进行词形还原。我注意到一些奇怪的事情： 并非所有单词都被词形还原。要说的是，

selected -> select

哪个是对的。

然而，involved !-> involve and horsing !-> horse除非我明确输入“v”（动词）属性。

在 python 终端上，我得到了正确的输出，但在我的终端中却没有得到正确的输出code http://pastebin.com/midgUJ02:

>>> from nltk.stem import WordNetLemmatizer
>>> from nltk.corpus import wordnet
>>> lem = WordNetLemmatizer()
>>> lem.lemmatize('involved','v')
u'involve'
>>> lem.lemmatize('horsing','v')
u'horse'

代码的相关部分是这样的：

for l in LDA_Row[0].split('+'):
    w=str(l.split('*')[1])
    word=lmtzr.lemmatize(w)
    wordv=lmtzr.lemmatize(w,'v')
    print wordv, word
    # if word is not wordv:
    #   print word, wordv

整个代码是here http://pastebin.com/midgUJ02.

问题是什么？

如果您使用默认设置，则词形还原器需要正确的 POS 标签才能准确WordNetLemmatizer.lemmatize()，默认标签是名词，参见https://github.com/nltk/nltk/blob/develop/nltk/stem/wordnet.py#L39 https://github.com/nltk/nltk/blob/develop/nltk/stem/wordnet.py#L39

要解决该问题，请始终在词形还原之前对数据进行 POS 标记，例如

>>> from nltk.stem import WordNetLemmatizer
>>> from nltk import pos_tag, word_tokenize
>>> wnl = WordNetLemmatizer()
>>> sent = 'This is a foo bar sentence'
>>> pos_tag(word_tokenize(sent))
[('This', 'DT'), ('is', 'VBZ'), ('a', 'DT'), ('foo', 'NN'), ('bar', 'NN'), ('sentence', 'NN')]
>>> for word, tag in pos_tag(word_tokenize(sent)):
...     wntag = tag[0].lower()
...     wntag = wntag if wntag in ['a', 'r', 'n', 'v'] else None
...     if not wntag:
...             lemma = word
...     else:
...             lemma = wnl.lemmatize(word, wntag)
...     print lemma
... 
This
be
a
foo
bar
sentence

请注意“is -> be”，即

>>> wnl.lemmatize('is')
'is'
>>> wnl.lemmatize('is', 'v')
u'be'

用你的例子中的词来回答这个问题：

>>> sent = 'These sentences involves some horsing around'
>>> for word, tag in pos_tag(word_tokenize(sent)):
...     wntag = tag[0].lower()
...     wntag = wntag if wntag in ['a', 'r', 'n', 'v'] else None
...     lemma = wnl.lemmatize(word, wntag) if wntag else word
...     print lemma
... 
These
sentence
involve
some
horse
around

请注意，WordNetLemmatizer 有一些怪癖：

• python 中的 wordnet 词形还原和词性标记 https://stackoverflow.com/questions/15586721/wordnet-lemmatization-and-pos-tagging-in-python
• Python NLTK 使用 wordnet 对单词“further”进行词形化 https://stackoverflow.com/questions/22999273/python-nltk-lemmatization-of-the-word-further-with-wordnet

此外，NLTK 的默认词性标注器正在进行一些重大更改以提高准确性：

• Python NLTK pos_tag 未返回正确的词性标记 https://stackoverflow.com/questions/30821188/python-nltk-pos-tag-not-returning-the-correct-part-of-speech-tag
• https://github.com/nltk/nltk/issues/1110 https://github.com/nltk/nltk/issues/1110
• https://github.com/nltk/nltk/pull/1143 https://github.com/nltk/nltk/pull/1143

对于词形还原器的开箱即用/现成解决方案，您可以看看https://github.com/alvations/pywsd https://github.com/alvations/pywsd以及我如何进行一些 try- excepts 来捕获 WordNet 中不存在的单词，请参阅https://github.com/alvations/pywsd/blob/master/pywsd/utils.py#L66 https://github.com/alvations/pywsd/blob/master/pywsd/utils.py#L66