Hotel_Reviews_Classifier

Classifies hotel reviews as True/Fake and Positive/Negative The word tokens in the reviews are used as features for the classification.

Classifiers:

1. Naive Bayes Classifier
2. Perceptron Classifier

• nblearn - learns a naive bayes model from the data
• nbclassify - classifies the reviews and assigns labels as true/fake and positive/negative

Smoothing was performed to deal with unknown vocabulary in the test data

• perceplearn - learns vanilla and averaged models from the data
• percepclassify - uses the vanilla and averaged models to classify the data