SYAHRIN, MUHAMMAD ALVIN (2025) ANALISIS SENTIMEN MASYARAKAT TWITTER TERHADAP JUDI ONLINE DI INDONESIA MENGGUNAKAN METODE INDOROBERTA DAN SVM TF-IDF. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
01 COVER.pdf Download (475kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
02 BAB 1.pdf Restricted to Registered users only Download (104kB) |
|
|
Text (BAB II)
03 BAB 2.pdf Restricted to Registered users only Download (259kB) |
|
|
Text (BAB III)
04 BAB 3.pdf Restricted to Registered users only Download (213kB) |
|
|
Text (BAB IV)
05 BAB 4.pdf Restricted to Registered users only Download (1MB) |
|
|
Text (BAB V)
06 BAB 5.pdf Restricted to Registered users only Download (31kB) |
|
|
Text (DAFTAR PUSTAKA)
07 DAFTAR PUSTAKA.pdf Restricted to Registered users only Download (112kB) |
|
|
Text (LAMPIRAN)
08 LAMPIRAN.pdf Restricted to Registered users only Download (1MB) |
Abstract
The rapid advancement of information and communication technology has fueled the widespread use of social media as a platform for public expression, including opinions on social issues such as online gambling. This study aims to analyze public sentiment in Indonesia on Twitter regarding the phenomenon of online gambling using two algorithmic approaches: IndoRoBERTa and SVM with TF-IDF. The dataset consists of 11,841 Indonesian-language tweets collected over a two-month period. The tweets were categorized into three sentiment classes: positive, neutral, and negative. This research implements the IndoRoBERTa model as a deep learning approach based on the transformer architecture, and SVM with TF-IDF as a conventional machine learning method. Evaluation results show that the IndoRoBERTa model achieved an accuracy of 81.27%, while the SVM TF-IDF model achieved 79.74%. Despite class imbalance in the dataset where negative sentiment dominates at 55.6% IndoRoBERTa demonstrated more accurate and stable performance in sentiment classification. The findings indicate that the majority of Indonesian users express negative perceptions toward online gambling. The study also recommends the application of data balancing techniques, expansion of data sources, and exploration of hybrid model approaches for future research. Kata kunci: Online Gambling, Sentiment Analysis, Twitter, NLP, IndoRoBERTa, SVM, TF-IDF Perkembangan teknologi informasi dan komunikasi telah mendorong maraknya aktivitas media sosial sebagai wadah ekspresi masyarakat, termasuk dalam menyuarakan opini terkait isu-isu sosial seperti judi online. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat Indonesia di platform Twitter terhadap fenomena judi online dengan menggunakan dua pendekatan algoritma: IndoRoBERTa dan SVM TF-IDF. Dataset yang digunakan terdiri dari 16.899 tweet berbahasa Indonesia yang dikumpulkan selama periode dua bulan. Data tersebut diklasifikasikan ke dalam tiga kategori sentimen, yaitu positif, netral, dan negatif. Penelitian ini mengimplementasikan model IndoRoBERTa sebagai representasi dari pendekatan deep learning berbasis transformer, serta SVM dengan representasi fitur TF-IDF sebagai pendekatan machine learning konvensional. Hasil evaluasi menunjukkan bahwa model IndoRoBERTa mencapai akurasi sebesar 81,27%, sedangkan SVM TF-IDF memperoleh akurasi sebesar 79,74%. Meskipun terdapat ketidakseimbangan kelas pada dataset dengan dominasi sentimen negatif sebesar 55,6% model IndoRoBERTa mampu mengklasifikasikan sentimen secara lebih akurat dan stabil. Temuan penelitian mengungkapkan bahwa mayoritas masyarakat Indonesia memiliki persepsi negatif terhadap judi online. Penelitian ini juga merekomendasikan penggunaan teknik penyeimbangan data, perluasan sumber dataset, serta eksplorasi pendekatan model hibrida dalam penelitian selanjutnya. Kata kunci: Judi Online, Analisis Sentimen, Twitter, NLP, IndoRoBERTa, SVM, TF-IDF
Actions (login required)
 |
View Item |
