Analisis Tren dan Kebaruan Pendekatan Hybrid Transformer untuk Meningkatkan Akurasi Deteksi Hoaks Berbahasa Indonesia
DOI:
https://doi.org/10.33998/jakakom.2026.6.1.2873Kata Kunci:
Deteksi Hoaks, NLP, Deep learning, Transformer Models, Hybrid ApproachAbstrak
Penelitian ini bertujuan untuk menganalisis perkembangan metodologis, kesenjangan penelitian, serta kebaruan pendekatan dalam teknik deteksi hoaks berbahasa Indonesia yang semakin kompleks seiring meningkatnya penyebaran misinformasi digital. Permasalahan utama yang diangkat adalah keterbatasan model konvensional dalam memahami konteks linguistik, ketidakseimbangan data, serta rendahnya interpretabilitas model. Penelitian ini menggunakan metode literature review non-systematic terhadap 31 artikel ilmiah terindeks nasional dan internasional periode 2019–2025 dengan pendekatan analisis tematik dan komparatif. Hasil penelitian menunjukkan adanya peningkatan signifikan performa model dari pendekatan machine learning klasik dengan akurasi 85–97%, ke deep learning sebesar 90–99%, hingga model transformer-based seperti IndoBERT dan RoBERTa yang mencapai akurasi 94–99,5%. Pendekatan hybrid seperti Bi-LSTM + IndoBERT dan CNN-LSTM menunjukkan performa paling stabil dengan akurasi hingga 99,58% serta peningkatan generalisasi model. Selain itu, teknik augmentasi data seperti back-translation terbukti meningkatkan akurasi secara signifikan hingga lebih dari 5%. Namun demikian, masih ditemukan kesenjangan pada aspek interpretabilitas, efisiensi komputasi, serta integrasi multimodal. Kesimpulannya, pendekatan hybrid berbasis transformer merupakan solusi paling efektif dalam meningkatkan akurasi dan stabilitas deteksi hoaks, meskipun masih diperlukan pengembangan pada aspek Explainable AI dan multimodalitas untuk meningkatkan keandalan sistem di dunia nyata.
Unduhan
Referensi
Yakub, Sy. Y., Wowor, R. J., Agustriawan, D., Irmawati, Kristiyanti, D. A., Winarno, P. M., Hassan, R., Mohamad, R., & Razzaq, R. Z. (2025). Deep learning for cyber security: Deephoax detection. International Conference on Artificial Intelligence and Soft Computing.
Rifai, A. P., Mulyani, Y. P., Febrianto, R., Arini, H. M., Wijayanto, T., Lathifah, N., Liu, X., Li, J., Yin, H., Wu, Y., & Mohawesh, R. (2023). DETECTION MODEL FOR FAKE NEWS ON COVID-19 IN INDONESIA. None.
Tandiano, A. H., & Jollyta, D. (2024). CLASSIFICATION OF FAKE NEWS IN INDONESIAN LANGUAGE USING SUPPORT VECTOR MACHINE METHOD. None.
Jollyta, D., Gusrianty, G., Prihandoko, P., & Sukrianto, D. (2023). N-gram and kernel performance using Support Vector Machine algorithm for fake news detection system. Ilkom Jurnal Ilmiah.
Rohman, M. A., Khairani, D., Hulliyah, K., Arini, Riswandi, P., & Lakoni, I. (2021). Systematic literature review on methods used in classification and fake news detection in indonesian. None.
Syaifuddiin, G. N., Arifin, R., Desriyanti, D., Buntoro, G. A., Rosyidin, Z. U., Pratama, R. Y., & Selamat, A. (2023). Hoax identification of indonesian tweeters using ensemble classifier. The Academic Center for Education, Culture and Research.
Hendrawan, R., Hana, K. M., & Kurniati, A. (2024). Enhancing indonesian abusive language detection on imbalanced news comment dataset using Support Vector Machine with oversampling. None.
Yusuf, R., & Suyanto, S. (2022). Hoax detection on indonesian text using long short-term memory. None.
Mashuri, C., Prastyo, E. H. A., & Hariri, F. R. (2025). Improving fake news detection accuracy with lexicon-based approach and LSTM through text preprocessing. JURNAL SISTEM INFORMASI BISNIS.
Hati, C. R. S., & Sulistiani, H. (2025). Implementation of CNN algorithm for indonesian hoax news detection on online news portals. JOURNAL OF APPLIED INFORMATICS AND COMPUTING.
Sibaroni, Y., Mahadzir, S., Prasetiyowati, S., & Ihsan, A. F. (2024). Combating misinformation: Leveraging deep learning for hoax detection in indonesian political social media. Jurnal Infotel.
Fathin, M., Sibaroni, Y., & Prasetyowati, S. S. (2024). Handling imbalance dataset on hoax indonesian political news classification using IndoBERT and random sampling. None.
Jocelynne, C., Wijayakusuma, I. L., & Harini, L. P. I. (2025). Detection of political hoax news using fine-tuning IndoBERT. JOURNAL OF APPLIED INFORMATICS AND COMPUTING.
Jin, R., Fu, R., Wen, Z., Zhang, S., Liu, Y., & Tao, J. (2024). Fake news detection and manipulation reasoning via large vision-language models. arXiv.org.
Mohawesh, R., Maqsood, S., & Althebyan, Q. (2023). Multilingual deep learning framework for fake news detection using capsule neural network. Journal of Intelligence and Information Systems.
Syari, M. A., Sembiring, H., & Siregar, M. F. (2025). Integration of critical thinking in intelligent algorithms for hoax detection on social media platforms. None.
Laimeheriwa, J., Iswanto, I. A., Oktriani, W., & Hidayat, M. F. (2024). A survey on indonesian hoax analyzer and fake news detection using deep learning techniques. None.
Kozik, R., Pawlicka, A., Pawlicki, M., Chora, M., Mazurczyk, W., & Cabaj, K. (2024). A meta-analysis of state-of-the-art automated fake news detection methods. IEEE Transactions on Computational Social Systems.
Wibowo, F. W., Dahlan, A., & Wihayati. (2021). Detection of fake news and hoaxes on information from web scraping using classifier methods. None.
Noor, A. Z. M., Gernowo, R., & Nurhayati, O. D. (2023). Data Augmentation for hoax detection through the method of convolutional neural network in indonesian news. None.
Fariha, T. M. A., & Lhaksmana, K. (2025). Indonesian fake news classification using bi-LSTM with IndoBERT and Data Augmentation. International Conference Advancement Data Science, E-Learning and Information Systems.
Li, X., Zhang, Y., & Malthouse, E. (2024). Large language model agent for fake news detection. arXiv.org.
Kurniawan, E. M., Aherman, V. N., Qomariyah, N. N., Manuaba, I., & Anom, A. (2023). Analysing hoax dataset in indonesian language with topic modeling. International Conferences on Information Science and System.
Ferdiansyah, I. F., & Wella. (2022). Comparative study of hoax detection using Support Vector Machine algorithm, naive bayes, random forest and k-nearest neighbor. Information and Communication Technology Convergence.
Rachmawati, O. C. R., & Darmawan, Z. M. E. (2024). The comparison of deep learning models for indonesian political hoax news detection. Journal.
Yefferson, D. Y., Lawijaya, V., & Girsang, A. S. (2024). Hybrid model: IndoBERT and long short-term memory for detecting indonesian hoax news. IAES International Journal of Artificial Intelligence (IJ-AI).
Purnamadewi, Y. R., & Zahra, A. (2025). Enhancing detection of zero-day phishing email attacks in the indonesian language using deep learning algorithms. Bulletin of Electrical Engineering and Informatics.
Adiati, N. P. R., Priambodo, D. F., Girinoto, G., Indarjani, S., Rizal, A., Prayoga, A., & Beatrix, Y. (2024). Comparative study of predictive models for hoax and disinformation detection in indonesian news. International Journal of Advances in Intelligent Informatics.
Rozi, I. F., Arianto, R., & Mahdyan, H. H. (2023). Fake news detection using sentiment analysis approach in indonesian language. None.
Mulyawan, R., Naparin, H., & Fatihia, W. M., “Comparison of Text Vectorization Methods for IMDB Movie Review Sentiment Analysis Using SVM,” Journal of Applied Informatics and Computing, 2025.
Mulyawan, R., Sunyoto, A., & Muhammad, A. H., "Automatic Indonesian Image Captioning using CNN and Transformer-based Model Approach," 2022 5th International Conference on Information and Communications Technology (ICOIACT), Yogyakarta, Indonesia, 2022.
