Question

我正在使用句子标记器，但如何从输出中删除不需要的/ n

from nltk.tokenize import word_tokenize, sent_tokenize
from nltk.corpus import stopwords
import PyPDF2 as p2
pdf_file = open("Muhammad_CV.pdf", 'rb')
pdf_read = p2.PdfFileReader(pdf_file)
count = pdf_read.numPages

for i in range(count):
    page = pdf_read.getPage(i)
    text = page.extractText()                               #Extract text
    tokenized = sent_tokenize(text)                 #Token
    all_words = []
    for w in tokenized:
    all_words.append(w.lower())                  #Lower case
# ///////////////// Stop Words ///////////////////////////
    stop_words = set(stopwords.words('english'))
    filtered = []
    for w in all_words:
        if w not in stop_words:
        filtered.append(w)
    print(filtered)

我得到的输出：

{'the specialization includes:\n \n\n \nintroduction\n \nto\n \ndata\n \nscience\n \n\n \nbig\n \ndata\n \n&\n \ncloud\n \ncomputing\n \n\n \ndata\n \nmining\n \n\n \nmachine\n \nlearn\ning'}

所需的输出：

{'the specialization includes: introduction to data science big data cloud\n computing data mining machine learning'}

Answer 1

您只需要调用字符串strip()方法即可删除周围的空白。

这是一个示例（也使用了理解，因为这是pythonic的方法：））

from nltk.tokenize import word_tokenize, sent_tokenize
from nltk.corpus import stopwords
import PyPDF2 as p2
pdf_file = open("Muhammad_CV.pdf", 'rb')
pdf_read = p2.PdfFileReader(pdf_file)
count = pdf_read.numPages

for i in range(count):
    page = pdf_read.getPage(i)
    text = page.extractText()
    tokenized = sent_tokenize(text)
    all_words = [w.strip().lower() for w in tokenized]
    stop_words = set(stopwords.words('english'))
    filtered = [w for w in all_words if w not in stop_words]
    print(filtered)

编辑：将trim更正为strip：）

Answer 2

 text = '''\n Apple has quietly  hired Dr. Rajiv B. Kumar, a pediatric endocrinologist \n. He will continue working at the hospital part time \n '''

 tokenized_sent_before_remove_n = nltk.sent_tokenize(text)
 #op 
 ['\n Apple has quietly  hired Dr. Rajiv B. Kumar, a pediatric endocrinologist \n.',
'He will continue working at the hospital part time']


 tokenized_sent_after_remove_n = [x.replace('\n','') for x in tokenized_sent]
 #o/p
 [' Apple has quietly  hired Dr. Rajiv B. Kumar, a pediatric endocrinologist .',
 'He will continue working at the hospital part time']

如何使用nltk在使用send_tokenize的同时从输出屏幕中删除\ n

2 个答案: