I need to parse sentences from a paragraph in Python. Is there an existing package to do this, or should I be trying to use regex here?
Asked
Active
Viewed 2.3k times
13
-
Are there double-spaces after the end of each sentence? – Zach Young Feb 28 '12 at 00:07
-
2Your problem statement doesn't provide sufficient information for us to work with. – RanRag Feb 28 '12 at 00:10
-
2There are some answers here: http://stackoverflow.com/questions/116494/python-regular-expression-to-split-paragraphs – martineg Feb 28 '12 at 00:10
-
4"Purely syntactic approaches using regexps sound problematic... just think of the 5.5 ways that Prof. Smith from the U.S. told us periods can be used." – DSM Feb 28 '12 at 00:17
-
These things are usually done by dedicated sentence splitter tools / library modules. Trying to do with regexes alone is not going to produce good results. The better splitters have been trained. – tchrist Feb 28 '12 at 00:33
2 Answers
46
The nltk.tokenize
module is designed for this and handles edge cases. For example:
>>> from nltk import tokenize
>>> p = "Good morning Dr. Adams. The patient is waiting for you in room number 3."
>>> tokenize.sent_tokenize(p)
['Good morning Dr. Adams.', 'The patient is waiting for you in room number 3.']

strcat
- 5,376
- 1
- 27
- 31
0
Here is how I am getting the first n sentences:
def get_first_n_sentence(text, n):
endsentence = ".?!"
sentences = itertools.groupby(text, lambda x: any(x.endswith(punct) for punct in endsentence))
for number,(truth, sentence) in enumerate(sentences):
if truth:
first_n_sentences = previous+''.join(sentence).replace('\n',' ')
previous = ''.join(sentence)
if number>=2*n: break #
return first_n_sentences
Reference: http://www.daniweb.com/software-development/python/threads/303844

David542
- 104,438
- 178
- 489
- 842
-
it will not work if the text contains URL's, or something with punctuation (period) on the term like, Ms., Dr., etc. – mannysz Aug 03 '16 at 00:20