ANALISIS SENTIMEN TWEET BERBAHASA INDONESIA DI TWITTER
| Main Authors: | , PAULINA ALIANDU, , Drs. Edi Winarko, Ph.D |
|---|---|
| Format: | Thesis NonPeerReviewed |
| Terbitan: |
[Yogyakarta] : Universitas Gadjah Mada
, 2012
|
| Subjects: | |
| Online Access: |
https://repository.ugm.ac.id/99095/ http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=55674 |
Daftar Isi:
- Sentiment analysis is a computational study of opinions, sentiments and emotions expressed in text. Web from review online sites, blogs, social networks contain a large number of opinion sources for individual and organizations needs. One of social networking that contain opinion data is Twitter. With the facility that called Twitter API, Twitter be able provide recent tweets that posted by Twitter user. This recent tweets, can be accessed by query on particular term, in order to generate sentiment on particular query term. This research is aimed to build an application that can determine public sentiment on Indonesian tweet in Twitter by user query on particular object term. The method that being used in this research is Naive Bayes. The method was used to build classification model on training data. In order to make sentiment class anotation easier, emoticon has been used. Training data were collected using crontab by querying emoticon and national media accounts that linked to the Twitter API. The collected data will pass particular preprocessing before training on. Weighting features that being used are term frequency with laplace smoothing and TF-IDF. All of the data that being used in this research are tweet post in Indonesian. From the implementation results obtained an accuracy of 77,45% using term frequency with laplace smoothing and 75,86% using TF-IDF on test set that anotated by emoticons. The results of manually marked test set are 70,68% for term frequency with laplace smoothing and 71,26% for TF-IDF. Accuracy measurement also done by using Support Vector Machine in RapidMiner. The results obtained an accuracy of 77,79% using term frequency and 77,57% using TF-IDF. Accuracy that produced by Support Vector Machine Method is better than Naive Bayes Method.