CountVectorizer忽略了“我”

Question

为什么CountVectorizer在滑雪板中忽略代词"I"？

ngram_vectorizer = CountVectorizer(analyzer = "word", ngram_range = (2,2), min_df = 1)
ngram_vectorizer.fit_transform(['HE GAVE IT TO I'])
<1x3 sparse matrix of type '<class 'numpy.int64'>'
ngram_vectorizer.get_feature_names()
['gave it', 'he gave', 'it to']

Answer 1

默认标记器只考虑两个字符(或更多)字。

您可以通过将适当的token_pattern传递给CountVectorizer来更改此行为。

默认模式是(请参阅文档中的签名)：

'token_pattern': u'(?u)\\b\\w\\w+\\b'

例如，您可以通过更改默认值获得一个不删除一个字母单词的CountVectorizer：

from sklearn.feature_extraction.text import CountVectorizer
ngram_vectorizer = CountVectorizer(analyzer="word", ngram_range=(2,2), 
                                   token_pattern=u"(?u)\\b\\w+\\b",min_df=1)
ngram_vectorizer.fit_transform(['HE GAVE IT TO I'])
print(ngram_vectorizer.get_feature_names())

这意味着：

['gave it', 'he gave', 'it to', 'to i']