Improving Usual Naive Bayes Classifier Performances with Neural Naive Bayes based Models

Elie Azeraf, Elie Azeraf, Emmanuel Monfrini, Wojciech Pieczynski

2022

Abstract

Naive Bayes is a popular probabilistic model appreciated for its simplicity and interpretability. However, the usual form of the related classifier suffers from two significant problems. First, as caring about the observations’ law, it cannot consider complex features. Moreover, it considers the conditional independence of the observations given the hidden variable. This paper introduces the original Neural Naive Bayes, modeling the classifier’s parameters induced from the Naive Bayes with neural network functions. This method allows for correcting the first default. We also introduce new Neural Pooled Markov Chain models, alleviating the conditional independence assumption. We empirically study the benefits of these models for Sentiment Analysis, dividing the error rate of the usual classifier by 4.5 on the IMDB dataset with the FastText embedding, and achieving an equivalent F1 as RoBERTa on TweetEval emotion dataset, while being more than a thousand times faster for inference.

Paper Citation

in Harvard Style

Azeraf E., Monfrini E. and Pieczynski W. (2022). Improving Usual Naive Bayes Classifier Performances with Neural Naive Bayes based Models. In Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-549-4, pages 315-322. DOI: 10.5220/0010890400003122

in Bibtex Style

@conference{icpram22,
author={Elie Azeraf and Emmanuel Monfrini and Wojciech Pieczynski},
title={Improving Usual Naive Bayes Classifier Performances with Neural Naive Bayes based Models},
booktitle={Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2022},
pages={315-322},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010890400003122},
isbn={978-989-758-549-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 11th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Improving Usual Naive Bayes Classifier Performances with Neural Naive Bayes based Models
SN - 978-989-758-549-4
AU - Azeraf E.
AU - Monfrini E.
AU - Pieczynski W.
PY - 2022
SP - 315
EP - 322
DO - 10.5220/0010890400003122