The Development of Indonesian Sentiment Analysis with Negation Handling

DOI

10.2991/aisr.k.200424.010How to use a DOI?

Keywords

sentiment analysis, negation handling, Multinomial Naïve Bayes

Abstract

The polarity of a text can be seen based on the words used in the text. The use of negation words in a sentiment can change the polarity of the sentiment. One example of a negation word in Indonesian is the word ‘tidak’. If a sentiment has a word “tidak” combined with a word “jelek” to form “tidak jelek”, then the sentiment is classified into positive class. However, if the combination of the word “tidak jelek” is processed as a word that has no relationship, then the sentiment will be falsely classified into negative class. The existence of negation word in a sentiment is one of the challenges in sentiment analysis, because it can cause ambiguity and misclassification when classifying text based on its polarity. This research shows that negation handling cannot be done only by doing syntactic analysis, but it also requires semantic analysis to overcome the problem of ambiguity.

Copyright

© 2020, 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  - Novi YUSLIANI
AU  - Mastura Diana MARIESKA
AU  - Endang LESTARI
AU  - M. Ridho Putra SULFA
AU  - Wibi ARIMURTI
PY  - 2020
DA  - 2020/05/06
TI  - The Development of Indonesian Sentiment Analysis with Negation Handling
BT  - Proceedings of the Sriwijaya International Conference on Information Technology and Its Applications (SICONIAN 2019)
PB  - Atlantis Press
SP  - 67
EP  - 72
SN  - 1951-6851
UR  - https://doi.org/10.2991/aisr.k.200424.010
DO  - 10.2991/aisr.k.200424.010
ID  - YUSLIANI2020
ER  -