Fast and accurate sentiment classification using an enhanced Naive Bayes model

Vivek Narayanan, Ishan Arora, Arjun Bhatia
5/27/2013
cs.CLcs.IRcs.LG

Official code: vivekn/sentiment (506 stars, BSD-2-Clause, last commit 2020-10-01).

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.

View on arXivView PDF

Code Implementations(1)

vivekn/sentimentOfficial100%
BSD-2-Clause

Sentiment analysis using machine learning techniques.

506191PythonOct 19, 20126 years agoBSD-2-Clause

Similar Papers

From TF-IDF to Transformers: A Comparative and Ensemble Approach to Sentiment Classification

84% match2026

Natural Language Processing Through Transfer Learning: A Case Study on Sentiment Analysis

82% match2023

A Comparison of Lexicon-Based and ML-Based Sentiment Analysis: Are There Outlier Words?

82% match2023

Leveraging ChatGPT As Text Annotation Tool For Sentiment Analysis

82% match2023

Urdu Speech and Text Based Sentiment Analyzer

81% match2022

Classification of Public Opinion on the Free Nutritional Meal Program on YouTube Media Using the LSTM Method

81% match2026

Cite this paper

@article{narayanan2013fast,
  title  = {Fast and accurate sentiment classification using an enhanced Naive Bayes model},
  author = {Vivek Narayanan and Ishan Arora and Arjun Bhatia},
  year   = {2013},
  eprint = {1305.6143},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1305.6143}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1305.6143.svg)](https://codeofpaper.com/paper/1305.6143)

Discussion