The objective of this project is to classify the given reviews as either Positive or Negative. The project uses NLTK for data preprocessing. With the help of the TextBlob module we have found the polarity and subjectivity of the sentences. Logistic Regression is used to classify the given reviews as Positive or Negative with an accuracy of 88% and the model is tested on some unseen data with a testing accuracy of 86%
The dataset used contains 3000 reviews taken from imdb.com, yelp.com and amazon.com. The reviews contain labels which indicate whether it is Positive (1) or Negative (0)