This project implements a Naive Bayes classifier to classify restaurant reviews from Zomato into distinct sentiment categories (Good, Okay, Bad). The entire implementation is developed from scratch without utilizing any inbuilt machine learning libraries, focusing on the fundamental principles of the Naive Bayes algorithm and the application of probability theory in text classification tasks.
The dataset used for this project is the Zomato Reviews & Ratings from Kaggle. You can download it from the following link: Zomato Reviews & Ratings Dataset.
The accuraies as of now follows Validation Accuracy: 51%, Test Accuracy: 47%
I am trying to build more accurate models.