Implementation of Naive Bayes Classification Method for Sentiment Analysis on Community Opinion to Indonesian Criminal Code Draft
- DOI
- 10.2991/assehr.k.201010.027How to use a DOI?
- Keywords
- The draft law on criminal law, Text Mining, Sentiment Analysis, Classification
- Abstract
Current developments of technology greatly facilitate the public to access information by both print media and social media. One example of social media that is widely used by the public is Twitter, because users can also comment on issues that are being discussed. One of them is an issue related to the controversy article in the draft law on criminal law to cause demonstrations conducted by students in several regions in Indonesia. Many people who think, both positive and negative opinions obtained from twitter. The method used in this study is the Naive Bayes Classification method which is a classification method with a simple probability that applies the Bayes theorem with high (independent) assumption. The advantage of the Naive Bayes method is that this method has high speed and accuracy when applied in large databases and diverse data. From the analysis, it was obtained a total of 3561 tweet data consist of 30.27% positive sentiment and 69.73% negative sentiment. Then from the classification result obtained an accuracy of 93.12%, a recall of 99.20%, a precision of 91.65% and an area under curve (AUC) value of 89.08%, which means the classification is very good.
- 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/).
Cite this article
TY - CONF AU - Sheila Farach Diba AU - Jaka Nugraha PY - 2020 DA - 2020/10/11 TI - Implementation of Naive Bayes Classification Method for Sentiment Analysis on Community Opinion to Indonesian Criminal Code Draft BT - Proceedings of the 2nd International Seminar on Science and Technology (ISSTEC 2019) PB - Atlantis Press SP - 186 EP - 192 SN - 2352-5398 UR - https://doi.org/10.2991/assehr.k.201010.027 DO - 10.2991/assehr.k.201010.027 ID - Diba2020 ER -