Muhammad Rezka Al Maghribi, . (2026) IMPLEMENTASI RETRIEVAL-AUGMENTED GENERATION UNTUK SISTEM TANYA JAWAB AL-QUR’AN BERBASIS SIMILARITY SEARCH. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (697kB) |
|
Text
AWAL.pdf Download (3MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (4MB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (15MB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (8MB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (34MB) |
|
Text
BAB 5.pdf Download (1MB) |
|
Text
DAFTAR PUSTAKA.pdf Download (2MB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository staff only Download (129kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (5MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (29MB) |
|
Text
Artikel KI.pdf Restricted to Repository staff only Download (498kB) |
Abstract
Muslim university students and/or Muslim alumni who belong to the digital native group often seek digital explanations of the Qur'an, yet they still face difficulties in verifying source authority and accuracy, an initial survey showed that 50% of respondents experienced this problem. This study aims to design, develop, and evaluate a mobile-based Qur'anic question-answering system capable of providing accurate, understandable, and source-traceable answers. The research applies a software engineering approach using the Waterfall model through the stages of requirements analysis, design, implementation, and testing. The system is developed using a Retrieval-Augmented Generation (RAG) Architecture that integrates Flutter, FastAPI, Qdrant, and the official corpus of the Indonesian Ministry of Religious Affairs, while implementing E5-based semantic retrieval, Strict Grounding, an adaptive Similarity threshold, Negative rejection, and a cloud-based LLM fallback. Evaluation was conducted through backend and frontend Unit tests, Black-box Testing, PSSUQ, and expert validation. The results show that all 12 backend Unit tests and 12 frontend Unit tests passed, Black-box Testing reached 100%, the PSSUQ score was 1.53 (highly acceptable), and expert validation achieved an average of 3.8 out of 4. The system is considered feasible as a supporting medium for Qur'anic learning, although it remains limited to the Kemenag RI corpus and does not yet support a local LLM fallback.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210511086] [Pembimbing 1: Musthofa Galih Pradana] [Pembimbing 2: Radinal Setyadinsa] [Penguji 1: Ridwan Raafi’udin] [Penguji 2: Muhammad Panji Muslim] |
| Uncontrolled Keywords: | RAG, Question Answering Systems, Al-Qur'an, Semantic Search, hallucination |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science T Technology > T Technology (General) |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | MUHAMMAD REZKA AL MAGHRIBI |
| Date Deposited: | 03 Sep 2026 03:07 |
| Last Modified: | 03 Sep 2026 03:07 |
| URI: | http://repository.upnvj.ac.id/id/eprint/51652 |
Actions (login required)
 |
View Item |