METODE LEXICON-LEARNING BASED UNTUK IDENTIFIKASI TWEET OPINI BERBAHASA INDONESIA
| Main Author: | Azhar, Yufis |
|---|---|
| Format: | Article PeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
Prodi Pendidikan Teknik Informatika Jurusan Teknik Informatika Fakultas Teknik Dan Kejuruan Universitas Pendidikan Ganesha
, 2017
|
| Subjects: | |
| Online Access: |
http://eprints.umm.ac.id/58113/7/Azhar%20-%20Opinion%20Tweet%20Lexicon%20Based%20Learning%20Based.pdf http://eprints.umm.ac.id/58113/ https://ejournal.undiksha.ac.id/index.php/janapati/article/view/11739 |
Daftar Isi:
- Social media has long used the community to convey opinions and facts about an event, especially twitter. Many methods are proposed to extract tweets containing opinions. Among them using a keyword identification approach in a tweet better known as lexicon based. Although this method has a high enough precision value in identifying an opinion tweet, the resulting recall value is quite low. This is because of the limitations of lexicon used as identifier. In this research, proposed combination of lexicon based and machine learning method in optimizing the result of identification of opinion tweet. The test results showed a significant increase in recall value when compared with lexicon based method while maintaining precision value