Exercise

Sentiment analysis for movie reviews

In this exercise you'll explore the probabilities outputted by logistic regression on a subset of the Large Movie Review Dataset.

The variables X and y are already loaded into the environment. X contains features based on the number of times words appear in the movie reviews, and y contains labels for whether the review sentiment is positive (+1) or negative (-1).

Instructions

100 XP

Train a logistic regression model on the movie review data.
Predict the probabilities of negative vs. positive for the two given reviews.
Feel free to write your own reviews and get probabilities for those too!