Fast and accurate sentiment classification using an enhanced Naive Bayes model

TL;DR

This paper enhances the Naive Bayes classifier for sentiment analysis by combining negation handling, n-grams, and feature selection, achieving high accuracy with fast training and testing.

Contribution

It introduces an improved Naive Bayes approach that significantly boosts sentiment classification accuracy using simple, computationally efficient techniques.

Findings

Achieved 88.80% accuracy on IMDB dataset

Combined negation handling, n-grams, and mutual information for feature selection

Maintained linear training and testing time complexities

Abstract

We have explored different methods of improving the accuracy of a Naive Bayes classifier for sentiment analysis. We observed that a combination of methods like negation handling, word n-grams and feature selection by mutual information results in a significant improvement in accuracy. This implies that a highly accurate and fast sentiment classifier can be built using a simple Naive Bayes model that has linear training and testing time complexities. We achieved an accuracy of 88.80% on the popular IMDB movie reviews dataset.