Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Teknik Elektro dan Komputasi (ELKOM)
Sugiarto
Sekolah Tinggi Meteorologi Klimatologi dan Geofisika

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analisis Sentimen Komentar Tiktok Pasca Gempa Mindanao Dengan Pemodelan SVM-TFIDF dan IndoBERT-BiLSTM Reza Okta Pratama; Arfany Dhimas Muftareza; Nardi; Sugiarto
Jurnal Teknik Elektro dan Komputasi (ELKOM) Vol. 8 No. 2 (2026): Jurnal Teknik Elektro dan Komputasi (ELKOM)
Publisher : Universitas Muhammadiyah Jember

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32528/elkom.v8i2.6294

Abstract

This study analyzes the sentiment of Indonesian-language TikTok comments in response to the M7.7 Mindanao, Philippines earthquake (June 8, 2026), which caused 47 fatalities, hundreds of injuries, over 20,000 displaced persons, and tsunami waves felt along Indonesian coasts. A total of 2,294 comments were collected from six TikTok videos via web scraping using Apify, with 2,023 validated after preprocessing. Automatic labeling using Indonesian RoBERTa-base produced three classes: negative (49.23%), neutral (38.79%), and positive (12.08%). Two models were compared: SVM + TF-IDF with an RBF kernel and a hybrid IndoBERT + BiLSTM model. SVM + TF-IDF achieved 67.16% accuracy with a macro F1-score of 0.58 in 0.31 seconds, while IndoBERT + BiLSTM achieved 66.42% accuracy with a macro F1-score of 0.64 but required 2,249.36 seconds. IndoBERT + BiLSTM outperformed SVM on the minority class with a positive F1-score of 0.57 versus 0.38. Linguistic analysis revealed negative sentiment dominated by threat vocabulary, positive by religious expressions, and neutral by informational content.
Co-Authors Arfany Dhimas Muftareza Nardi Reza Okta Pratama