Implementation of IndoBERT for Sarcasm Detection in YouTube Comments on Economic Issues
Implementasi IndoBERT untuk Deteksi Sarkasme pada Komentar YouTube tentang Isu Ekonomi
DOI:
https://doi.org/10.21070/ups.11791Keywords:
IndoBERT, sarkasme, komentar YouTube, teks Indonesia, ekonomiAbstract
This study implements IndoBERT to detect sarcasm in Indonesian YouTube comments discussing economic issues. The data were collected from public comments on economic videos covering staple food prices, inflation, fuel prices, electricity tariffs, and purchasing power from 2023 to 2025. After crawling, filtering, manual labeling, and light text preprocessing, the valid dataset was split using an 80:20 stratified scheme. Random Oversampling was applied only to the training data to reduce class imbalance. The IndoBERT Base P1 model was then fine-tuned for binary classification into sarcastic and non-sarcastic comments. The test results show an accuracy of 85.87%, weighted precision of 85.12%, weighted recall of 85.87%, and weighted F1-score of 85.35%. These findings indicate that IndoBERT can capture contextual patterns in informal Indonesian comments and improve the identification of implicit criticism in public economic discourse.
Downloads
References
R. Putra Perssela, R. Mahendra, and W. Rahmadianti, “Pemanfaatan Media Sosial untuk Efektivitas Komunikasi,” Jurnal Ilmiah Mahasiswa Kuliah Kerja Nyata, vol. 2, no. 3, pp. 650–656, 2022.
S. Kemp, “Digital 2025: Indonesia,” DataReportal, 2025. [Online]. Available: https://datareportal.com/reports/digital-2025-indonesia
A. Saadillah, “Penggunaan Bahasa Sarkasme Netizen di Media Sosial,” vol. 9, no. 2, pp. 1437–1447, 2023.
M. Jia, C. Xie, and L. Jing, “Debiasing Multimodal Sarcasm Detection with Contrastive Learning,” Proc. AAAI Conf. Artificial Intelligence, vol. 38, no. 16, pp. 18354–18362, 2024.
N. Modami, E. E. R. Manopo, D. R. Enditama, and A. T. Ayunda, “Analisis Sentimen Komentar YouTube terhadap Rumor Peluncuran iPhone 17 Menggunakan Web Scraping dan Studi Komparatif Algoritma Klasifikasi,” JOISIE (Journal of Information Systems and Informatics Engineering), vol. 9, no. 2, 2025.
M. Muadin, Junadhi, Rahmiati, and H. Asnal, “Implementasi Metode Support Vector Machine pada Opinion Mining Masyarakat Terkait ChatGPT,” JOISIE (Journal of Information Systems and Informatics Engineering), vol. 7, no. 1, 2023.
D. Suhartono, W. Wongso, and A. T. Handoyo, “IdSarcasm: Benchmarking and Evaluating Language Models for Indonesian Sarcasm Detection,” IEEE Access, vol. 12, pp. 87323–87332, 2024.
S. P. Octarini, A. Y. Zakiyyah, and K. Purwandari, “Comparison of IndoBERT and SVM Algorithm to Perform Aspect Based Sentiment Analysis using Hierarchical Dirichlet Process,” vol. 7, no. 3, pp. 363–370, 2025.
D. R. Alfinsyah and B. P. Hartato, “Evaluating the Impact of Random Over Sampling on IndoBERT Performance for Indonesian Sentiment Analysis,” vol. 9, no. 6, pp. 3270–3282, 2025.
A. A. Ningtyas, A. Solichin, and R. Pradana, “Analisis Sentimen Komentar YouTube tentang Prediksi Resesi Ekonomi Tahun 2023 Menggunakan Algoritme Naive Bayes,” vol. 20, no. 1, pp. 9–16, 2023.
P. Paramita and A. Ibrahim, “Analisis Sentimen terhadap Pengguna QRIS (Quick Respond Code Indonesian Standart) pada Twitter Menggunakan Metode Naive Bayes Classifier,” JOISIE (Journal of Information Systems and Informatics Engineering), vol. 7, no. 1, 2023.
A. M. Fanani, “Sarcasm Detection in Indonesian YouTube Comments using Fine-Tuned IndoBERT with Class Imbalance Handling,” vol. 10, no. 1, pp. 26–36, 2026.
A. R. Hanum, I. A. Zetha, J. N. Fajrina, R. A. Wulandari, C. Putri, S. P. Andina, and N. Yudistira, “Performance Analysis of the BERT Text Classification Algorithm,” Jurnal Teknologi Informasi dan Ilmu Komputer, vol. 11, no. 3, pp. 537–546, 2024.
D. K. Sumartha, “Implementasi IndoBERT untuk Analisis Sentimen Opini Publik terhadap Kebijakan Kenaikan UKT di Era Pemerintahan,” Jurnal Informatika dan Teknik Elektro Terapan, vol. 13, no. 3S1, pp. 867–875, 2025.
B. Wilie et al., “IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language Understanding,” Proc. AACL-IJCNLP, pp. 843–857, 2020.
Downloads
Additional Files
Posted
License
Copyright (c) 2026 UMSIDA Preprints Server
This work is licensed under a Creative Commons Attribution 4.0 International License.
