1. Learn
    2. /
  2. Courses
    3. /
  3. Feature Engineering for NLP in Python

Exercise

Higher order n-grams for sentiment analysis

Similar to a previous exercise, we are going to build a classifier that can detect if the review of a particular movie is positive or negative. However, this time, we will use n-grams up to n=2 for the task.

The n-gram training reviews are available as X_train_ng. The corresponding test reviews are available as X_test_ng. Finally, use y_train and y_test to access the training and test sentiment classes respectively.

Instructions

100 XP
  • Define an instance of MultinomialNB. Name it clf_ng
  • Fit the classifier on X_train_ng and y_train.
  • Measure accuracy on X_test_ng and y_test the using score() method.