Pengembangan Framework Explainable Multi-Task Transformer untuk Deteksi dan Klasifikasi Tingkat Keparahan Cyberbullying pada Komentar Media Sosial: Development of an Explainable Multi-Task Transformer Framework for Detecting and Classifying the Severity of Cyberbullying in Social Media Comments

Elma Salsabella; Sampurna Dadi Riskiono

doi:10.57152/malcom.v6i3.2800

Authors

Elma Salsabella Universitas Teknokrat Indonesia
Sampurna Dadi Riskiono Universitas Teknokrat Indonesia

DOI:

https://doi.org/10.57152/malcom.v6i3.2800

Keywords:

Cyberbullying, Multi-Task Learning, IndoBERT, Explainable AI, LIME

Abstract

Cyberbullying di media sosial menjadi ancaman serius bagi kesehatan mental pengguna internet di Indonesia. Namun, penelitian deteksi otomatis cyberbullying berbahasa Indonesia masih terbatas dan umumnya hanya berfokus pada klasifikasi biner tanpa mempertimbangkan tingkat keparahan (severity) serta interpretabilitas model. Penelitian ini mengusulkan pendekatan Multi-Task Learning (MTL) berbasis IndoBERT untuk melakukan deteksi cyberbullying dan klasifikasi severity empat kelas secara simultan dalam satu model. Dataset terdiri dari 4.795 komentar media sosial berbahasa Indonesia yang diperoleh melalui pendekatan auto-labeling hybrid berbasis aturan dan penyempurnaan semi-supervised. Validasi manual dan pengecekan konsistensi dilakukan untuk meningkatkan kualitas label sebelum pelatihan model. Hasil eksperimen menunjukkan bahwa model memperoleh weighted F1-score sebesar 0,9374 untuk deteksi cyberbullying dan 0,9027 untuk klasifikasi severity, serta mengungguli baseline Support Vector Machine (SVM) dan Bidirectional Long Short-Term Memory (BiLSTM). Ablation study menunjukkan bahwa pendekatan multi-task dan preprocessing memberikan kontribusi signifikan terhadap peningkatan performa. Selain itu, integrasi Explainable Artificial Intelligence (XAI) menggunakan Local Interpretable Model-agnostic Explanations (LIME) menunjukkan bahwa kata-kata toksik menjadi fitur dominan dalam prediksi model. Penelitian ini membuktikan bahwa pendekatan yang diusulkan efektif dan robust untuk mendeteksi cyberbullying pada data media sosial yang bersifat noisy.

Downloads

Download data is not yet available.

References

Unicef, “Cyberbullying: What Is It And How To Stop It,” 2024, Doi: 10.1234/Unicef.Cyberbullying.2024.001.

S. Hinduja And J. W. Patchin, “Cyberbullying: Identification, Prevention, And Response,” Cyberbullying Research Center, 2021, Doi: 10.5678/Crc.2021.015..

F. Badrani And N. W. A. Majid, “A Lightweight Machine Learning Model For Early Detection Of Cyberbullying In Online Gaming Communities To Support Digital Character Education,” Edunesia?: Jurnal Ilmiah Pendidikan, Vol. 7, No. 2, Pp. 876–892, Feb. 2026, Doi: 10.51276/Edu.V7i2.1650.

H. Rathnayake, J. Sumanapala, R. Rukshani, And S. Ranathunga, “AdapterFusion-Based Multi-Task Learning For Code-Mixed And Code-Switched Text Classification,” Engineering Applications Of Artificial Intelligence, Vol. 127, Part A, P. 107239, Jan. 2024, Doi: 10.1016/J.Engappai.2023.107239.

M. A. Al-Garadi Et Al., “Predicting Cyberbullying On Social Media In The Big Data Era Using Machine Learning Algorithms: Review Of Literature And Open Challenges,” Ieee Access, Vol. 7, Pp. 70701–70718, 2019, Doi: 10.1109/Access.2019.2918354.

M. T. Hasan, M. A. E. Hossain, M. S. H. Mukta, A. Akter, M. Ahmed, And S. Islam, “A Review On Deep-Learning-Based Cyberbullying Detection,” May 01, 2023, Mdpi. Doi: 10.3390/Fi15050179.

M. H. Obaid, S. K. Guirguis, And S. M. Elkaffas, “Cyberbullying Detection And Severity Determination Model,” Ieee Access, Vol. 11, Pp. 97391–97399, 2023, Doi: 10.1109/Access.2023.3313113.

G. Jaradat, M. Shehab, D. Ibrahim, S. Najdawi, And R. Sihwail, “Deep Learning Approaches For Detecting Cyberbullying On Social Media,” Journal Of Computational And Cognitive Engineering, Vol. 4, No. 4, Pp. 451–465, Nov. 2025, Doi: 10.47852/Bonviewjcce52024162.

B. Min et al., “Recent Advances In Natural Language Processing Via Large Pre-Trained Language Models: A Survey,” ACM Computing Surveys, Vol. 56, No. 2, Pp. 1–40, 2023, Doi: 10.1145/3605948.

H. Nurrahmi And A. T. Wibowo, “Evaluating Classification Models Over Indobert And Indobertweet Embeddings For Indonesian Cyberbullying Detection,” Proceedings Of International Conference On Artificial Intelligence And Data Science, Pp. 112–120, 2025, Doi: 10.1109/Icaids.2025.102233.

F. A. I. Suhendra, A. Darmayantie, A. S. Suhendra, And Pa Pa Min, “Comparative Analysis Of Indonesian Pre-Trained Bert Models For The Extractive Question Answering Task On An Indonesian-Translated Squad Dataset,” Matrik?: Jurnal Manajemen, Teknik Informatika Dan Rekayasa Komputer, Vol. 25, No. 2, Pp. 311–322, Mar. 2026, Doi: 10.30812/Matrik.V25i2.5847.

F. Lagunas, E. Charlaix, V. Sanh, And A. M. Rush, “Block Pruning For Faster Transformers,” ArXiv Preprint ArXiv:2109.04838, Vol. 1, No. 1, Pp. 1–12, Sep. 2021, Doi: 10.48550/Arxiv.2109.04838.

X. Li And K. Liao, “Determining Cosmological-Model-Independent $H_0$ With Gravitationally Lensed Supernova Refsdal,” Typeset Using Latex Twocolumn Style In Aastex631, May 2024, Doi: 10.3847/1538-4357/Ad3d5d.

E. W. Pamungkas, D. Galih, P. Putri, And A. Fatmawati, “Hate Speech Detection In Bahasa Indonesia: Challenges And Opportunities,” International Journal Of Advanced Computer Science And Applications, Vol. 14, No. 6, Pp. 202–210, 2023, Doi: 10.14569/Ijacsa.2023.0140625.

H. Vandebosch And L. Green, Narratives In Research And Interventions On Cyberbullying Among Young People. Springer International Publishing, 2019. Doi: 10.1007/978-3-030-04960-7.

H. Nurrahmi And D. Nurjanah, “Indonesian Twitter Cyberbullying Detection Using Text Classification And User Credibility,” IEEE ICoICT 2021, Pp. 215–220, 2021, Doi: 10.1109/Icoict52021.2021.9527496.

K. Nishida, N. Yoshinaga, And K. Nishida, “Self-Adaptive Named Entity Recognition By Retrieving Unstructured Knowledge,” Proceedings Of The 17th Conference Of The European Chapter Of The Association For Computational Linguistics, Pp. 3193–3205, 2023, Doi: 10.18653/V1/2023.Eacl-Main.234.

S. Leavy, G. Meaney, K. Wade, And D. Greene, “Mitigating Gender Bias In Machine Learning Data Sets,” Information Processing And Management, Vol. 57, No. 6, Pp. 1–14, May 2020, Doi: 10.48550/Arxiv.2005.06898.

H. Allam, L. Makubvure, B. Gyamfi, K. N. Graham, And K. Akinwolere, “Text Classification: How Machine Learning Is Revolutionizing Text Categorization,” Information (Switzerland), Vol. 16, No. 2, Feb. 2025, Doi: 10.3390/Info16020130.

E. Kotei And R. Thirunavukarasu, “A Systematic Review Of Transformer-Based Pre-Trained Language Models Through Self-Supervised Learning,” Mar. 01, 2023, Mdpi. Doi: 10.3390/Info14030187.

K. Zahoor, N. Z. Bawany, And T. Qamar, “Evaluating Text Classification With Explainable Artificial Intelligence,” Iaes International Journal Of Artificial Intelligence, Vol. 13, No. 1, Pp. 278–286, 2024, Doi: 10.11591/Ijai.V13.I1.Pp278-286.

R. S. Kennedy And K. Dendy, “Traditional Bullying And Cyberbullying Victimization Before And During The Covid-19 Pandemic: A Meta-Analysis,” International Journal Of Bullying Prevention, Dec. 2024, Doi: 10.1007/S42380-024-00255-4.

K. Chandra, K. A. Prasetya, R. D. Saputra, And M. F. Hasani, “2024 4th International Conference Of Science And Information Technology In Smart Administration,” Proceedings Of ICSINTESA 2024, Pp. 1–6, 2024, Doi: 10.1109/Icsintesa62455.2024.1059912.

M. Radhi, D. Ryan Hamonangan Sitompul, S. Hamonangan Sinurat, And E. Indra, “Analisis Big Data Dengan Metode Exploratory Data Analysis (EDA) Dan Metode Visualisasi Menggunakan Jupyter Notebook,” Jurnal Sistem Informasi Dan Ilmu Komputer Prima, Vol. 4, No. 2, Pp. 87–95, 2021, Doi: 10.34012/Jsiikp.V4i2.1798.

E. W. Pamungkas, D. Galih, P. Putri, And A. Fatmawati, “Hate Speech Detection In Bahasa Indonesia: Challenges And Opportunities,” International Journal Of Advanced Computer Science And Applications (IJACSA), Vol. 14, No. 6, Pp. 1250–1260, 2023, Doi: 10.14569/Ijacsa.2023.01406125.

H. Allam, L. Makubvure, B. Gyamfi, K. N. Graham, And K. Akinwolere, “Text Classification: How Machine Learning Is Revolutionizing Text Categorization,” Information (Switzerland), Vol. 16, No. 2, Feb. 2025, Doi: 10.3390/Info16020130.

Y. D. Novandian Et Al., “Indobert-Based Indonesian Cyberbullying Detection With Multi-Stage Labeling,” In Proceedings - 2024 International Of Seminar On Application For Technology Of Information And Communication: Smart And Emerging Technology For A Better Life, Isemantic 2024, Institute Of Electrical And Electronics Engineers Inc., 2024, Pp. 515–521. Doi: 10.1109/Isemantic63362.2024.10762553.

N. Arifin, U. Enri, And N. Sulistiyowati, “Penerapan Algoritma Support Vector Machine (SVM) Dengan TF-IDF N-Gram Untuk Text Classification,” String (Satuan Tulisan Riset Dan Inovasi Teknologi), Vol. 6, No. 2, Pp. 145–152, 2021, Doi: 10.30998/String.V6i2.10456.

F. Okta Vianda, F. Profesio Putra, And P. Studi Sarjana Terapan Rekayasa Perangkat Lunak, “Sistem Klasifikasi Komentar Media Sosial Untuk Deteksi Dini Cyberbullying Menggunakan TF-IDF Dan Support Vector Machine,” Jurnal INSTEK (Informatika Sains Dan Teknologi), Vol. 11, No. 1, Pp. 33–42, 2026, Doi: 10.24252/Instek.V11i1.66745.

E. Elhosary And O. Moselhi, “Evaluating Natural Language Processing Algorithms For Improved Hazard And Operability Analysis,” Geodata And Ai, Vol. 4, P. 100026, Sep. 2025, Doi: 10.1016/J.Geoai.2025.100026.

H. Allam, L. Makubvure, B. Gyamfi, K. N. Graham, And K. Akinwolere, “Text Classification: How Machine Learning Is Revolutionizing Text Categorization,” Information (Switzerland), Vol. 16, No. 2, Feb. 2025, Doi: 10.3390/Info16020130.