Analisis Sentimen dan Pemodelan Topik Terhadap Uji Coba Vaksin Tuberkulosis M72/AS01E di Indonesia Menggunakan Model Berbasis IndoBERT dan Topic Modeling

Santoso, Trista Avrilia (2026) Analisis Sentimen dan Pemodelan Topik Terhadap Uji Coba Vaksin Tuberkulosis M72/AS01E di Indonesia Menggunakan Model Berbasis IndoBERT dan Topic Modeling. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5026221008-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB)

Abstract

Indonesia menempati peringkat kedua sebagai negara dengan total kasus Tuberkulosis (TBC) terbanyak di dunia menurut World Health Organization Global Tuberculosis Report tahun 2023. Persetujuan kerja sama pemerintah untuk melaksanakan uji coba vaksin TBC M72/AS01E di Indonesia memicu kontroversi pro dan kontra di masyarakat. Tugas Akhir ini bertujuan untuk menganalisis sentimen masyarakat dan mengidentifikasi topik utama yang dibahas terhadap isu uji coba vaksin melalui kolom komentar di platform YouTube. Model analisis sentimen dibangun dengan membandingkan tiga algoritma, yaitu IndoBERT + Dense, IndoBERT + GRU, dan IndoBERT + LSTM. Model pemodelan topik dibangun dengan membandingkan tiga algoritma, yaitu Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), dan BERTopic. IndoBERT + GRU terpilih sebagai model terbaik untuk analisis sentimen dengan F1-Score mencapai 93.22%. Distribusi opini publik terdiri atas 49,9% sentimen negatif, 40,2% sentimen netral, dan 9,9% sentimen positif. BERTopic terpilih sebagai model terbaik untuk pemodelan topik sentimen negatif, sedangkan LDA terpilih sebagai model terbaik untuk pemodelan topik sentimen positif dan sentimen netral. Sentimen negatif terdiri atas empat topik utama yaitu isu objek penelitian, isu motif program, risiko kesehatan, dan teori konspirasi. Sentimen netral terdiri atas dua topik utama yaitu prosedur pelaksanaan dan fakta kasus di Indonesia. Sentimen positif terdiri atas dua topik utama yaitu optimisme program dan harapan publik. Tugas Akhir ini menyimpulkan bahwa mayoritas masyarakat menolak program uji coba vaksin TBC M72/AS01E. Temuan ini memberikan saran strategis bagi pemerintah untuk memperjelas tujuan, prosedur, dan standar keamanan uji coba vaksin TBC M72/AS01E.
================================================================================================================================
Indonesia ranks second as the country with the highest cases of Tuberculosis (TBC) in the world, according to the World Health Organization Global Tuberculosis Report in 2023. The government’s approval to conduct the clinical trial of the tuberculosis M72/AS01E vaccine in Indonesia has caused public pros and cons. This thesis aims to analyze public sentiment and identify the main topics discussed regarding the vaccine trial issue through the comment sections on the YouTube platform. The sentiment analysis model was developed by comparing three algorithms: IndoBERT + Dense, IndoBERT + GRU, and IndoBERT + LSTM. The topic modeling model was conducted by comparing three algorithms: Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), and BERTopic. IndoBERT + GRU was selected as the best-performing sentiment analysis model by achieving an F1-Score of 93.22%. Public opinion was distributed by 49.9% negative sentiment, 40.2% neutral sentiment, and 9.9% positive sentiment. BERTopic was selected as the best model for negative sentiment topic modeling, while LDA was selected for both positive and neutral sentiment topic modeling. Negative sentiment consisted four main topics, namely research subject concerns, program motive issues, health risks, and conspiracy theories. Neutral sentiment consisted two main topics, namely implementation procedures and TB case facts in Indonesia. Positive sentiment consisted two main topics, namely program optimism and public hope. This thesis finds that the majority of the public opposes the M72/AS01E TB vaccine trial program. These findings suggest strategic recommendations for the government to better communicate the objectives, procedures, and safety standards of the TB vaccine trial to the public.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Vaksin, TBC, Analisis Sentimen, Pemodelan Topik, IndoBERT Vaccine, TBC, Sentiment Analysis, Topic Modeling, IndoBERT
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Information Technology > Information System > 57201-(S1) Undergraduate Thesis
Depositing User:	Trista Avrilia Santoso
Date Deposited:	28 Jul 2026 03:56
Last Modified:	28 Jul 2026 03:56
URI:	http://repository.its.ac.id/id/eprint/138341

Actions (login required)

View Item