Muhammad Fadawkas Oemarki, . (2026) PENGEMBANGAN SISTEM KONSULTASI DAN REKOMENDASI TINDAK LANJUT HUKUM PIDANA BERBASIS WEBSITE DENGAN PENERAPAN RAG DAN LLM. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (139kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (184kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (407kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (734kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (9MB) |
|
Text
BAB 5.pdf Download (146kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (152kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (164kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (28MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (4MB) |
Abstract
Limited access to accurate and consistent criminal law information remains a significant issue due to unstructured legal content that is difficult for non-legal users to comprehend. This study aims to design and develop an artificial intelligence–based criminal law consultation system utilizing Large Language Models and the Retrieval-Augmented Generation approach. The system is designed to generate responses that are relevant to user queries and supported by criminal law documents as the underlying knowledge base. The research methodology includes system requirements analysis, collection and processing of criminal law documents into a vector database, and implementation of the RAG architecture. Model performance was evaluated using RAGAS metrics, achieving a faithfulness score of 0.407, answer relevancy of 0.801, context precision of 1.00, and context recall of 0.663. These results indicate that the system generates highly relevant answers with very precise contextual grounding, although further improvements are required to expand contextual coverage and enhance answer faithfulness to the source documents. Functional testing through black-box testing achieved a 100% success rate, while User Acceptance Testing resulted in an average score of 4.6 out of 5, indicating that the system is feasible for use as a web-based criminal law consultation platform.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No. Panggil: 2210511116] [Pembimbing 1: Neny Rosmawarni] [Pembimbing 2: Muhammad Adrezo] [Penguji 1: Tjahjanto] [Penguji 2: Kharisma Wiati Gusti |
| Uncontrolled Keywords: | Artificial Intelligence, Criminal Law Consultation, LLM, RAG, Legal Information System |
| Subjects: | Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | MUHAMMAD FADAWKAS OEMARKI |
| Date Deposited: | 08 Feb 2026 01:46 |
| Last Modified: | 08 Feb 2026 01:46 |
| URI: | http://repository.upnvj.ac.id/id/eprint/42510 |
Actions (login required)
 |
View Item |