1

i want to do a sentiment analysis on twitter tweet and for that i have test data set in csv file in which tweet column as input but how to create a training data and how to label the data and how to use as feature set.

please correct me if I'm wrong here.. i have to manually create the feature set and label each word?

till now i have done word tokenize and stop word and porter stemming to my test dataset. now from here i want apply my training set and as output i want positive and negative tweet result. please give me some correct direction..

swati
  • 11
  • 1
  • 3
  • [Learning to Classify Text.](http://www.nltk.org/book/ch06.html) – alexis May 18 '17 at 19:16
  • See http://stackoverflow.com/questions/21107075/classification-using-movie-review-corpus-in-nltk-python and http://stackoverflow.com/questions/29275614/using-my-own-corpus-instead-of-movie-reviews-corpus-for-classification-in-nltk – alvas May 19 '17 at 00:58
  • @alvas thank you for the links.. it's really help me. – swati May 25 '17 at 14:12
  • now i have another problem i.e. i don't have any labelled data. i want to train my test data against labelled train dataset. so, how i can make a labelled train dataset (label data should consist of negative and positive sentence/ words)? – swati May 25 '17 at 14:15

0 Answers0