A Comparison of Naïve Bayes and Bayesian Network on The Classification of Hijaiyah Pronunciation with Punctuation Letters

DOI

10.2991/icoiese-18.2019.9How to use a DOI?

Keywords

bayesian network; hijaiyah pronunciation; naïve bayes

Abstract

Arabic is a unique language because it really concerns in makhraj (the way of sound is made) that differentiate letters and words. The difference in pronouncing letters and words make the meaning of those words different, because pronunciation in Qur’an letters really concern in harakat (the length of words). According to that matter, it is necessary to build a speech recognition for Hijaiyah with punctuation letters in Qur’an. There are many methods that can be used for building that system. One of the best method is Hidden Markov Model (HMM). Main inference in HMM is Bayes’ Rule. Bayes’ Rule also used in Naïve Bayes, a part of Bayesian Network. This paper focused on Naïve Bayes and Bayesian Network. Before recognizing the data, first the data will be pre-processed using Linear Predictive Coding (LPC) for extracting cepstral coefficient that will be used as input in classifier. This system give a best micro average F1 score result, 76,67%, with Bayesian Network.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Adiwijaya Adiwijaya
AU  - Annisa Riyani
AU  - Mohamad Syahrul Mubarok
PY  - 2019/03
DA  - 2019/03
TI  - A Comparison of Naïve Bayes and Bayesian Network on The Classification of Hijaiyah Pronunciation with Punctuation Letters
BT  - Proceedings of the 2018 International Conference on Industrial Enterprise and System Engineering (IcoIESE 2018)
PB  - Atlantis Press
SP  - 48
EP  - 52
SN  - 2589-4943
UR  - https://doi.org/10.2991/icoiese-18.2019.9
DO  - 10.2991/icoiese-18.2019.9
ID  - Adiwijaya2019/03
ER  -