PRASETYO, PANJI AL MUQSITH (2024) TWITTER SENTIMENT ANALYSIS ON THE 2024 PRESIDENTIAL ELECTION USING THE NAÏVE BAYES MULTINOMIAL ALGORITHM. Tugas Akhir thesis, Informatics.
![[img]](http://eprints.uty.ac.id/style/images/fileicons/text.png) |
Text
5200411249_PANJI AL MUQSITH PRASETYO_Abstrak.pdf Download (13kB) |
Abstract
ABSTRACT The presidential election is a form of public enthusiasm in choosing a new leader who is worthy of leading the Indonesian state. Generally, people choose presidential and vice presidential candidates based on their educational background, experience in leading a region, and achievements achieved in leading. However, quite a few people choose based on promises and academic ability. Before the presidential election was held in 2019, many opinions from the internet community, especially the Indonesian people, often attacked and criticized the arguments of the presidential and vice presidential candidates, both in the context of promises made during the leadership period and attacking arguments based on personal reasons. The aim of developing this sentiment analysis application is that internet users who are still unfamiliar with the language can be helped in sorting information. Another aim is to find out whether a Naïve Bayes method or algorithm can run on website-based applications. In Indonesia, citizenship education and Indonesian language literacy are still very limited in frontier, remote and underdeveloped areas. This is a problem if the Indonesian people cannot understand the context of a comment that is circulating in Indonesia. It would be easier for the public if there was a system that runs specifically to understand someone's statement. With these problems, a system is needed to analyze website-based sentiment. The framework used is Flask with the Python programming language and with the help of the Naïve Bayes library in its implementation. With 540 data generated from the crawling process, the data will be tested under several available conditions, including data sharing and random state differences. This research produces an accuracy value of 77% with precision of 55.56%%, recall of 55.56%%, and F1-score of 53.33%, where these results use a case test_size of 0.1 which makes it the case with the best output in model building. Keywords: Presidential Election, Sentiment Analysis, Naïve Bayes Classifier.
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 10 Aug 2024 04:18 |
| Last Modified: | 10 Aug 2024 04:18 |
| URI: | http://eprints.uty.ac.id/id/eprint/15992 |
Actions (login required)
 |
View Item |
