Chatbot Fiqih Berbasis Retrieval-Augmented Generation pada Kitab Safinatun Najah
DOI:
https://doi.org/10.29407/fq3wvf38Abstract
Abstrak— Perkembangan teknologi kecerdasan buatan mendorong pemanfaatan chatbot sebagai media penyedia informasi keagamaan berbasis digital. Namun, penggunaan Large Language Model (LLM) masih menghadapi permasalahan berupa hallucination, yaitu kondisi ketika model menghasilkan informasi yang tidak sesuai dengan sumber rujukan. Penelitian ini bertujuan mengembangkan chatbot fiqih berbasis metode Retrieval-Augmented Generation (RAG) dengan Kitab Safinatun Najah sebagai sumber pengetahuan utama. Sistem dikembangkan menggunakan model ADDIE (Analysis, Design, Development, Implementation, Evaluation) dan diimplementasikan menggunakan Streamlit, LangChain, ChromaDB, model embedding multilingual-e5-large, serta Gemini sebagai Large Language Model. Metode RAG diterapkan dengan menggabungkan proses retrieval dokumen relevan dan proses generasi jawaban sehingga respons yang dihasilkan tetap mengacu pada isi kitab. Evaluasi sistem difokuskan pada kinerja proses retrieval menggunakan 250 data pengujian yang berasal dari 50 pertanyaan utama dengan lima variasi pertanyaan pada setiap data. Hasil pengujian menunjukkan nilai retrieval accuracy, precision, recall, dan F1-score sebesar 76,80%, serta average similarity sebesar 0,7644. Hasil penelitian menunjukkan bahwa metode RAG mampu memahami berbagai variasi pertanyaan pengguna dan menemukan informasi yang relevan berdasarkan isi Kitab Safinatun Najah. Dengan demikian, pendekatan RAG berpotensi meningkatkan kualitas sistem tanya jawab fiqih sekaligus mengurangi risiko hallucination pada model generatif.
Keywords:
Chatbot, Retrieval-Augmented Generation, Safinatun Najah, Fiqih, Kecerdasan Buatan.##plugins.themes.default.displayStats.downloads##
References
[1] W. Fan et al., "A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models," Proc. ACM SIGKDD Int. Conf. Knowl. Discov. Data Min., pp. 6491-6501, 2024, doi: 10.1145/3637528.3671470.
[2] M. Alshammary, M. N. Uddin, and L. Khan, "RFPG: Question-Answering from Low-Resource Language (Arabic) Texts using Factually Aware RAG," Proc. 2024 IEEE 10th Int. Conf. Collab. Internet Comput. CIC 2024, no. Cic, pp. 107-116, 2024, doi: 10.1109/CIC62241.200023.
[3] G. Lang and T. Gurpinar, "AI-Powered Learning Support: A Study of Retrieval-Augmented Generation (RAG) Chatbot Effectiveness in an Online Course," Inf. Syst. Educ. J., vol. 23, no. 2, pp. 4-13, 2025, doi: 10.62273/zklk5988.
[4] O. Nacar and A. Koubaa, "Enhancing Semantic Similarity Understanding in Arabic NLP with Nested Embedding Learning," Stud. Comput. Intell., vol. 1214, pp. 179-216, 2025, doi: 10.1007/978-3-031-90573-5_6.
[5] S. Syukran, "Strategi Pembelajaran Kitab Kuning dalam Pendidikan Agama Islam di Era Digital: Studi Kasus di Pondok Pesantren Darul Hijrah Al-Aziziyah," Edu Soc. J. Pendidikan, Ilmu Sos. Dan Pengabdi. Kpd. Masy., vol. 5, no. 2, pp. 2095-2108, 2025, doi: 10.56832/edu.v5i2.1688.
[6] T. Q. Ramadhani, N. Q. Nada, and N. D. S, "Penerapan Metode Retrieval-Augmented Generation (RAG) Pada Chatbot E-Commerce Berbasis Gemini Ai," J. Ilm. Ilk. - Ilmu Komput. Inform., vol. 8, no. 2, pp. 301-313, 2025, doi: 10.47324/ilkominfo.v8i2.384.
[7] A. Mishra and N. Brahmanapally, "A Comparative Performance Analysis of Locally Deployed Large Language Models Through a Retrieval-Augmented Generation Educational Assistant Application for Textual Data Extraction," AI, vol. 6, no. 6, pp. 0-26, 2025, doi: 10.3390/ai6060119.
[8] M. Renze and E. Guven, "The Effect of Sampling Temperature on Problem Solving in Large Language Models," EMNLP 2024 2024 Conf. Empir. Methods Nat. Lang. Process. Find. EMNLP 2024, pp. 7346-7356, 2024, doi: 10.18653/v1/2024.findings-emnlp.432.
[9] L. Huang et al., "A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions," ACM Trans. Inf. Syst., vol. 43, no. 2, pp. 1-58, 2025, doi: 10.1145/3703155.
[10] W. Zhang and J. Zhang, "Hallucination Mitigation for Retrieval-Augmented Large Language Models: A Review," Mathematics, vol. 13, no. 5, 2025, doi: 10.3390/math13050856.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work’s authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal’s published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License
