Analisa Sentimen Menggunakan Naïve Bayes Untuk Melihat Persepsi Masyarakat Terhadap Kenaikan Harga Jual Rokok Pada Media Sosial Twitter
| Main Authors: | Afshoh, Fauziah, -, Endang Wahyu Pamungkas, S.Kom, M.Kom. |
|---|---|
| Format: | Karya Ilmiah NonPeerReviewed application/pdf |
| Bahasa: | eng |
| Terbitan: |
, 2017
|
| Subjects: | |
| Online Access: |
http://eprints.ums.ac.id/49444/2/SURAT%20PERNYATAAN%20FAUZIAH%20AFSHOH%20%28L200130034%29.pdf http://eprints.ums.ac.id/49444/3/NASKAH%20PUBLIKASI%20FAUZIAH%20AFSHOH%20%28L200130034%29.pdf http://eprints.ums.ac.id/49444/ |
Daftar Isi:
- Social media Twitter is one example of social media that allows people to interact with each other. Twitter provides services to its users to send and read tweets that have been shared, so people prefer to pour of their opinions through social media rather than pass them directly. Public opinion contained in social media twitter be a perception, whether it is positive or negative. The huge amount of public opinion can be used as research material to locate information. Utilization of such information requires proper analysis techniques so that the resulting information can help some parties to take a decision. The use of the techniques in data processing can be completed using sentiment analysis or opinion mining. Therefore, in this study tries to analyze sentiment to see the public perception of the increase in cigarette prices on social media twitter using Naïve Bayes classifier to classify sentiment becomes positive, negative and neutral. The results of research that has been done can be seen that most of the positive sentiment was formed in response to the discourse of the increase in cigarette prices. In addition, results from testing the performance of the system using training data 150 positive, 150 negative and 50 neutral with Naïve Bayes classifier method produces a value classification accuracy better than using methods Lexicon Based.