Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Sistemasi: Jurnal Sistem Informasi KREATIF: Jurnal Pengabdian Masyarakat Nusantara
Oppir Hutapea
Institut Teknologi Del
Computer Science & IT Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Health Professions Law, Crime, Criminology & Criminal Justice

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Pengembangan Sistem Informasi Manajemen Administrasi dan Keuangan Berbasis Web Jemaat HKBP Palmarum Tarutung Ardiles Sinaga; Ana Muliyana; Oppir Hutapea
KREATIF: Jurnal Pengabdian Masyarakat Nusantara Vol. 5 No. 2 (2025): Jurnal Pengabdian Masyarakat Nusantara
Publisher : AMIK Veteran Purwokerto

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/kreatif.v5i2.6049

Abstract

The HKBP Palmarum Church faces significant challenges in its administration and financial management, which have been conducted manually. This approach exposes the system to risks of data loss, delays in information updates, and a lack of transparency in financial reporting. This community service project aims to develop a web-based management information system to address these issues. The methodology employed is the Software Development Life Cycle (SDLC) with a prototyping approach. The outcome of the project includes a system capable of managing congregation data, finances, and service information digitally. With the implementation of this system, church services are expected to become more efficient, transparent, and serve as a model for other churches.
Comparative Analysis of Knowledge Internalization through Supervised Fine-Tuning and Contextual Reasoning through Retrieval-Augmented Fine-Tuning in Large Language Models for Indonesian Village Regulations Oppir Hutapea; Syahrial Jeremia Sinaga; Vivaldi Adventus Simangunsong; Yohanssen Pratama; Adli Abdillah Nababan
SISTEMASI Vol 15, No 9 (2026): Sistemasi: Jurnal Sistem Informasi
Publisher : Universitas Islam Indragiri

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32520/stmsi.v15i9.6913

Abstract

Large language models (LLMs) are prone to hallucination, making their reliability in Indonesian-language legal question answering uncertain, particularly for Village Regulations, which are predominantly available as scanned PDF documents. This study provides a controlled comparison between knowledge internalization through Supervised Fine-Tuning (SFT) and contextual reasoning through Retrieval-Augmented Fine-Tuning (RAFT), using both LoRA and full fine-tuning from the same Llama-3.1-8B-Instruct checkpoint. The study also develops a traceable document-processing pipeline and a question-answer dataset derived from 55 Village Regulations from Badung Regency, consisting of 50 scanned and five born-digital documents. Model performance was evaluated using ROUGE-L, METEOR, BERTScore, and RAGAS. At the fourth epoch, RAFT-LoRA achieved the highest aggregate score (0.883), followed by RAFT-Full (0.881), while the best-performing SFT configuration reached 0.749. RAFT improved faithfulness from 0.640 to 0.872 and context recall from 0.637 to 0.907. The results indicate that training data structure had a greater impact on performance than the parameter-update strategy, while LoRA maintained RAFT performance with lower adaptation costs. These findings suggest that legal knowledge can be maintained in updatable external documents without requiring model retraining, while the model can be trained to generate evidence-grounded responses. However, the findings are limited to a single regency and automated evaluation; therefore, validation by legal experts is required before deployment.
Co-Authors Adli Abdillah Nababan Ana Muliyana Ardiles Sinaga Syahrial Jeremia Sinaga Vivaldi Adventus Simangunsong Yohanssen Pratama