Chatbot Hybrid Fatwa MUI Menggunakan Retrieval Augmented Generation dan Large Language Model
DOI:
https://doi.org/10.52060/juptik.v4i1.4318Abstrak
Aksesibilitas dokumen digital Fatwa Majelis Ulama Indonesia (MUI) yang terfragmentasi membuat pencarian informasi kurang efektif. Di sisi lain, sistem tanya jawab AI berbasis satu sumber dokumen (single-corpus) rentan menghasilkan jawaban tidak akurat (halusinasi) pada pertanyaan di luar domain. Penelitian ini mengembangkan Chatbot Hybrid Fatwa MUI menggunakan arsitektur Hybrid Retrieval bertingkat dengan dua sumber pengetahuan: dokumen Fatwa MUI sebagai korpus utama dan 12.370 hadis Bukhari-Muslim sebagai mekanisme cadangan (fallback). Sistem ini menerapkan pencarian semantik, verifikasi topik otomatis oleh model bahasa, dan pengalihan ke basis data hadis jika konteks fatwa dinilai tidak relevan. Hasil evaluasi menunjukkan peningkatan kesamaan makna jawaban sebesar 13,23% (dari 0,6664 menjadi 0,7546) dan peningkatan kesetiaan pada rujukan (faithfulness) sebesar 10,57% (dari 85,37% menjadi 94,39%), dengan tingkat penolakan (abstain rate) identik sebesar 26,83%. Pendekatan multi-korpus ini terbukti signifikan meningkatkan relevansi dan keakuratan jawaban dibandingkan RAG standar.
Referensi
[1] W. X. Zhao et al., “A Survey of Large Language Models,” arXiv Prepr. arXiv2303.18223, pp. 1–144, 2026, doi: https://doi.org/10.48550/arXiv.2303.18223.
[2] L. Xu, L. Sanders, K. Li, and J. C. L. Chow, “Chatbot for Health Care and Oncology Applications Using Artificial Intelligence and Machine Learning: Systematic Review,” JMIR Cancer, vol. 7, no. 4, 2021, https://doi.org/10.2196/27850
[3] M. I. Ardimansyah and M. H. Widianto, “Development of online learning media based on Telegram Chatbot (Case studies: Programming courses),” J. Phys. Conf. Ser., vol. 1987, no. 1, 2021, https://doi.org/10.1088/1742-6596/1987/1/012006
[4] M. Hasanuddin, M. S. Bin Sharuddin, and J. W. Mahri, “Pemodelan Fatwa Ekonomi Syariah dan Karakteristiknya di Indonesia,” Asy-Syari’ah, Vol, vol. 25, no. 1, pp. 1–16, 2023, https://doi.org/10.15575/as.v25i1.25373
[5] M. Shuhufi and F. Muhammad, “I Fatwas and Sharia-Based Policy Governance in Islamic Education Management in Indonesia,” vol. 10, no. 1, pp. 184–200, 2026, doi: 10.22373/sjhk.v10.i1.28762.
[6] R. R. Dewi, “Problematika Artificial Intelligence Sebagai Pemberi Fatwa Dalam Perspektif Hukum Islam,” J. Anal. Huk., vol. 7, no. 2, pp. 209–223, 2024, https://doi.org/10.38043/jah.v7i2.5137
[7] S. A. Putra, “AD-DUSTUR Jurnal Hukum dan Konstitusi AD-DUSTUR Jurnal Hukum dan Konstitusi,” AD-Dustur J. Huk. dan Konstitusi, vol. 1, no. 1, pp. 1–17, 2024. https://doi.org/10.58326/jad.v1i1.195
[8] S. M. T. I. Tonmoy, S. M. M. Zaman, A. Rani, V. Rawte, A. Chadha, and A. Das, “A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Models,” arXiv Prepr. arXiv2401.01313, 2023, doi: https://doi.org/10.48550/arXiv.2401.01313.
[9] Y. Gao et al., “Retrieval-Augmented Generation for Large Language Models : A Survey,” pp. 1–21, doi: https://doi.org/10.48550/arXiv.2312.10997.
[10] A. Asai, Z. Wu, Y. Wang, A. Sil, and H. Hajishirzi, “SELF-RAG: LEARNING TO RETRIEVE, GENERATE, AND CRITIQUE THROUGH SELF-REFLECTION,” arXiv Prepr. arXiv2509.01081, no. Step 1, 2023, doi: https://doi.org/10.48550/arXiv.2310.11511.
[11] M. Erlangga Fauzi and Tata Sutabri, “PublicTalk: Sistem Chatbot Pintar Berbasis Natural Language Processing untuk Layanan Pemerintahan Digital,” J. Sains Student Res., vol. 3, no. 2, pp. 426–433, 2025, https://doi.org/10.61722/jssr.v3i2.4325
[12] F. L. D. Cahyanti and R. D. A. Raya, “Perancangan Sistem Informasi Chatbot Retrieval Augmented Generation Berbasis Website Pada PT. Revolusi Cita Edukasi,” Indones. J. Comput. Sci., vol. 4, no. 1, pp. 15–21, 2025, https://doi.org/10.31294/m75d4782
[13] S. Rahayu, N. S. Harahap, S. Agustian, and P. Pizaini, “Penerapan Teknologi LangChain pada Question Answering System Fikih Empat Madzhab,” MALCOM Indones. J. Mach. Learn. Comput. Sci., vol. 4, no. 3, pp. 974–983, 2024, https://doi.org/10.57152/malcom.v4i3.1397
[14] T. Helviansyah, N. S. Harahap, M. Irsyad, and B. S. Negara, “Sistem Tanya Jawab Berbasis Chatbot Website Menggunakan Gemini Ai Pada Data Fiqih Kontemporer,” J. Inf. Syst. Manag., vol. 7, no. 1, pp. 38–47, 2025, https://doi.org/10.24076/joism.2025v7i1.2082
[15] N. A. M. Herwanza, N. S. Harahap, F. Yanto, and F. Insani, “Penerapan Langchain Retriever dengan Model Chat Openai dalam Pengembangan Sistem Chatbot Hadis Berbasis Telegram,” JTIM J. Teknol. Inf. dan Multimed., vol. 6, no. 1, pp. 70–83, 2024, https://doi.org/10.35746/jtim.v6i1.514
[16] F. I. Haekal, R. Setya Perdana, and P. P. Adikara, “Sistem Tanya Jawab Closed-Domain terhadap Dokumen Fatwa menggunakan Retrieval Augmented Generation dan Large Language Model,” vol. 9, no. 5, pp. 2548–964, 2025, [Online]. Available: http://j-ptiik.ub.ac.id
[17] A. Bouchekif, S. Rashwani, H. Sbahi, S. Gaben, M. Al-khatib, and M. Ghaly, “Assessing Large Language Models on Islamic Legal Reasoning: Evidence from Inheritance Law Evaluation,” 2025, doi: https://doi.org/10.48550/arXiv.2509.01081.
[18] L. Machado, R. Miyaji, R. Moulin, and S. Monc, “Evaluating RAG-based QA Systems : A Comparative Analysis of LLM as a Judge , Traditional Metrics , and Human Alignment,” 2024.
[19] F. Dobslaw, R. Feldt, J. Yoon, and S. Yoo, “Challenges in Testing Large Language Model Based Software : A Faceted Taxonomy,” rXiv Prepr. arXiv2503.00481, vol. 1, no. 1, pp. 1–20, 2025, doi: https://doi.org/10.48550/arXiv.2503.00481.
[20] L. Team and A. I. Meta, “The Llama 3 Herd of Models,” arXiv Prepr. arXiv2407.21783, pp. 1–92, 2024, doi: https://doi.org/10.48550/arXiv.2407.21783.
| Keywords | : |
Kata Kunci:
chatbot hybrid, MUI fatwa, retrieval-augmented generation (RAG), large language model (LLM), hybrid retrieval
|
| Galleys | : | |
| Diterbitkan | : |
2026-06-01
|
| Terbitan | : |
Hak Cipta (c) 2026 Jurnal Pengembangan Teknologi Informasi dan Komunikasi (JUPTIK)
Artikel ini berlisensi Creative Commons Attribution 4.0 International License.
