The goal was to explore logistic regression and feature engineering with existing Turi Create functions. In this notebook I have used product review data from Amazon.com to predict whether the sentiments about a product (from its reviews) are positive or negative.
-
Used SFrames to do some feature engineering
-
Trained a logistic regression model to predict the sentiment of product reviews.
-
Inspected the weights (coefficients) of a trained logistic regression model.
-
Made a prediction (both class and probability) of sentiment for a new product review.
-
Given the logistic regression weights, predictors and ground truth labels, wrote a function to compute the accuracy of the model.
-
Inspected the coefficients of the logistic regression model and interpreted their meanings.
-
Compared multiple logistic regression models.