如何检查Python中的列表中是否存在DataFrame字符串列的第一个单词？

Question

我有一个DataFrame df_sentences和一个列表question_words，如下所示：

df_sentences:

sentence                         label
you will not forget this movie   0
will the novel ever die          1
why we drink alcohol             1
did trump win the election       1
ambiance is perfect              0


question_words = ['what', 'why', 'when', 'where', 'whose', 'which', 'whom', 'who', 'how', 
                         'do', 'are', 'will', 'did', 'will', 'am', 'are', 'was', 'were', 'can', 'has', 'have']

我想检查sentence列中的第一个单词是否存在于列表question_words中，并将结果返回到新列ques_word中。

预期输出：

sentence                         label  ques_word
you will not forget this movie   0      0
will the novel ever die          1      1
why we drink alcohol             1      1
did trump win the election       1      1
the ambiance is perfect          0      0

到目前为止，我尝试使用的是.str.contains('|'.join(question_words)).astype(int)，但是正如预期的那样，它返回与question_words列表匹配的所有子字符串的所有数目。

Answer 1

.str.split(" ")[0].contains('|'.join(question_words)).astype(int)

应该做的事

Answer 2

如果需要快速解决方案，请使用列表理解。

q_set = set(question_words)
df['ques_word'] = [
    1 if w.split(None, 1)[0]  in q_set else 0 for w in df.sentence
]

df
                         sentence  label  ques_word
0  you will not forget this movie      0          0
1         will the novel ever die      1          1
2            why we drink alcohol      1          1
3      did trump win the election      1          1
4             ambiance is perfect      0          0