运行NLTK代码时触发TypeError: expected string or bytes-like object求助

嘿，我一眼就揪出问题所在了——你在循环里犯了个很容易踩的小坑：

在for i in range(len(sentence)):这个循环里，i是索引数字（比如0），但nltk.word_tokenize()需要的是字符串类型的句子内容，不是整数。你把索引值传给了分词函数，自然会触发TypeError: expected string or bytes-like object这个错误。

修正后的完整代码 #

只需要把words = nltk.word_tokenize(i)改成words = nltk.word_tokenize(sentence[i])就搞定了，完整代码如下：

import nltk
from nltk.stem import PorterStemmer
from nltk.corpus import stopwords

# 第一次运行务必下载必要的NLTK数据集，否则会报错
nltk.download('punkt')
nltk.download('stopwords')

paragraph = ''' State-run Bharat Sanchar Nigam Ltd (BSNL) is readying to pay November salary in another two days, which will be raised from internal accruals and bank loans.'''
sentence = nltk.sent_tokenize(paragraph)
stemmer = PorterStemmer()

for i in range(len(sentence)):
    # 这里要传入句子本身，而非循环索引i
    words = nltk.word_tokenize(sentence[i])
    words = [stemmer.stem(word) for word in words if word not in set(stopwords.words('english'))]
    sentence[i] = ' '.join(words)

# 可以打印看看处理后的结果
print(sentence)

额外优化小建议 #

其实你可以用更Pythonic的写法，直接遍历句子列表，不用手动处理索引，代码会更简洁易读：

processed_sentences = []
for sent in sentence:
    words = nltk.word_tokenize(sent)
    words = [stemmer.stem(word) for word in words if word not in set(stopwords.words('english'))]
    processed_sentences.append(' '.join(words))

内容的提问来源于stack exchange，提问作者jain