A Natural Language Processing Project.
The dataset containing reviews of customers of a restaurant where each review is classified as either positive review(1) or negative review(0) is used to train our ML model
Feature extraction techniques used: Bag of Words and TFIDF
the accuracy achieved using NaiveBayes algorithm:67%
The accuracy achieved using Support Vector Machine algorithm: on training set:99.7% on testing set:84%