0

I want to extract questions from question paper.I'm labeling each question as q and other sentences as i in dataset.

e.g.

Why is this sector becoming important in India,q
Describe any five public facilities needed for the development of a country,q
CBSE 10th English 2017 Unsolved Paper,i
THE MARKS ARE MENTIONED ON EACH QUESTION,i

Is it suitable to use textblob naivebayes classifier for this or should I use sklearn with nltk?

R. T
  • 11
  • 1
  • 7
  • Take a look at https://www.kaggle.com/alvations/basic-nlp-with-nltk – alvas Feb 05 '18 at 04:09
  • I'm not getting how to use basic nltk for extracting question. – R. T Mar 04 '18 at 06:05
  • Have you walkthrough all the code in the kaggle kernel? Also have you read the http://www.nltk.org/book/? – alvas Mar 04 '18 at 08:45
  • yes,still i'm not getting how to use nltk – R. T Mar 06 '18 at 17:16
  • Are you sure you went through the book? And the kaggle kernel? The kaggle kernel teaches you exactly how to build a classifier using sklearn and nltk for almost any classification task. – alvas Mar 07 '18 at 00:51
  • thanks for the links.It was helpful.Can you suggest something for https://stackoverflow.com/questions/49238091/is-it-correct-to-use-pos-tag-as-feature-in-questions-identification – R. T Mar 12 '18 at 14:59

0 Answers0