PERBANDINGAN MODEL SVM, BI-LSTM, DAN INDOBERTWEET UNTUK ANALISIS SENTIMEN KOMENTAR TWITTER PADA POSTINGAN PEMIMPIN NEGARA

BATISTUTA, MUHAMMAD ILZA (2026) PERBANDINGAN MODEL SVM, BI-LSTM, DAN INDOBERTWEET UNTUK ANALISIS SENTIMEN KOMENTAR TWITTER PADA POSTINGAN PEMIMPIN NEGARA. S1 thesis, Universitas Mercu Buana Jakarta.

Preview	Text (HAL COVER) Cover.pdf Download (739kB) | Preview
	Text (BAB I) Bab 1.pdf Restricted to Registered users only Download (59kB)
	Text (BAB II) Bab 2.pdf Restricted to Registered users only Download (162kB)
	Text (BAB III) Bab 3.pdf Restricted to Registered users only Download (137kB)
	Text (BAB IV) Bab 4.pdf Restricted to Registered users only Download (568kB)
	Text (BAB V) Bab 5.pdf Restricted to Registered users only Download (88kB)
	Text (DAFTAR PUSTAKA) Daftar Pustaka.pdf Restricted to Registered users only Download (94kB)
	Text (LAMPIRAN) Lampiran.pdf Restricted to Registered users only Download (537kB)

Abstract

This study compares SVM, Bi-LSTM, and IndoBERTweet for sentiment classification of Twitter/X comments on posts by a national leader for automatic sentiment labeling. The initial dataset consisted of 62,614 comments collected through GraphQL-based crawling. After preprocessing, 46,038 clean comments were obtained, while 10,000 comments were selected for manual labeling into three classes: positive, negative, and neutral. After preprocessing, 8,790 labeled data were used for model training, validation, and testing. Evaluation was conducted using accuracy, precision, recall, and F1-score. The results show IndoBERTweet achieved the best performance, with an accuracy of 0.7983 and a macro F1-score of 0.7612. The best model was used to automatically label the clean dataset. The automatic labeling results show that negative sentiment dominated with 25,100 comments, followed by neutral with 11,488 comments and positive with 9,450 comments. Monthly analysis indicates that negative comments peaked in September 2024 and were analyzed using word frequency, bigram, and wordcloud. Kata kunci: Sentiment Analysis, Twitter/X, SVM, Bi-LSTM, IndoBERTweet. Penelitian ini membandingkan SVM, Bi-LSTM, dan IndoBERTweet untuk klasifikasi sentimen komentar Twitter/X terhadap postingan pemimpin negara sebagai dasar pelabelan otomatis. Dataset awal terdiri atas 62.614 komentar yang dikumpulkan melalui crawling berbasis GraphQL. Setelah preprocessing, diperoleh 46.038 data bersih, sedangkan 10.000 data dipilih untuk pelabelan manual ke dalam tiga kelas, yaitu positif, negatif, dan netral. Setelah preprocessing, 8.790 data berlabel digunakan untuk pelatihan, validasi, dan pengujian model. Evaluasi dilakukan menggunakan accuracy, precision, recall, dan F1-score. Hasil penelitian menunjukkan bahwa IndoBERTweet memperoleh performa terbaik dengan accuracy 0,7983 dan macro F1-score 0,7612. Model terbaik tersebut digunakan untuk pelabelan otomatis pada data bersih. Hasil pelabelan menunjukkan sentimen negatif mendominasi sebanyak 25.100 komentar, diikuti netral 11.488 dan positif 9.450 komentar. Analisis bulanan menunjukkan komentar negatif tertinggi terjadi pada September 2024 dan dianalisis menggunakan word frequency, bigram, serta wordcloud. Temuan ini juga memberikan gambaran mengenai pola respons masyarakat terhadap postingan pemimpin negara di Twitter/X secara umum. Kata kunci: Analisis Sentimen, Twitter/X, SVM, Bi-LSTM, IndoBERTweet

Item Type:	Thesis (S1)
NIM/NIDN Creators:	41522010161
Uncontrolled Keywords:	Analisis Sentimen, Twitter/X, SVM, Bi-LSTM, IndoBERTweet
Subjects:	000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 004 Data Processing, Computer Science/Pemrosesan Data, Ilmu Komputer, Teknik Informatika 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 006 Special Computer Methods/Metode Komputer Tertentu > 006.3 Artificial Intelligence/Kecerdasan Buatan > 006.31 Machine Learning/Pembelajaran Mesin 000 Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 000. Computer Science, Information and General Works/Ilmu Komputer, Informasi, dan Karya Umum > 006 Special Computer Methods/Metode Komputer Tertentu > 006.7 Multimedia Systems/Sistem-sistem Multimedia > 006.75 Social Multimedia/Multimedia Social
Divisions:	Fakultas Ilmu Komputer > Informatika
Depositing User:	khalimah
Date Deposited:	27 Aug 2026 03:04
Last Modified:	27 Aug 2026 03:04
URI:	http://repository.mercubuana.ac.id/id/eprint/103447

Actions (login required)

View Item