Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Nusantara Hasana Journal Journal of Artificial Intelligence and Digital Business
Thamrin, Musdalifa
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Humanities Computer Science & IT Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Nursing

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Analisis Perbandingan Efektivitas Zero-Shot vs Chain-of-Thought Prompting dalam Meningkatkan Presisi Informasi pada LLM Berbasis Dokumen PDF Yahya, Kurnia; Moeis, Dikwan; Thamrin, Musdalifa; Yunarti, Sry; Felicia Watratan, Alvina; Mallu, Satriawaty
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 5 No. 1 (2026): Februari - April
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v5i1.6524

Abstract

Pemanfaatan Large Language Models (LLM) dalam ekosistem pendidikan tinggi untuk ekstraksi informasi akademik sering kali terkendala oleh fenomena halusinasi data yang dapat menyesatkan pengguna. Penelitian ini bertujuan untuk menganalisis perbandingan efektivitas antara teknik Zero-Shot dan Chain-of-Thought (CoT) prompting dalam upaya meningkatkan presisi informasi pada sistem Retrieval-Augmented Generation (RAG) yang berbasis dokumen PDF. Metode penelitian yang digunakan adalah eksperimen laboratorium dengan mengintegrasikan framework LangChain dan database vektor FAISS untuk memproses dokumen teknis berupa Panduan Akademik institusi. Evaluasi kualitas jawaban dilakukan melalui pendekatan hibrida, menggunakan metrik otomatis BERTScore untuk mengukur kemiripan semantik dan penilaian manusia (Human Evaluation) oleh pakar Teknologi Informasi untuk mengukur tingkat presisi serta validitas informasi secara kualitatif. Hasil penelitian menunjukkan bahwa teknik Chain-of-Thought secara konsisten mengungguli Zero-Shot di seluruh parameter evaluasi yang diuji. Peningkatan paling signifikan tercatat pada aspek validitas jawaban berdasarkan penilaian pakar sebesar 37,14%, serta kenaikan skor presisi pada metrik BERTScore sebesar 9,03%. Temuan ini membuktikan bahwa mekanisme penalaran logis secara bertahap pada teknik CoT mampu mereduksi halusinasi secara efektif dengan memastikan setiap klaim jawaban memiliki jejak audit yang kuat pada dokumen sumber. Implikasi praktis dari penelitian ini memberikan rekomendasi strategis bagi pengembang sistem informasi akademik untuk menerapkan pendekatan 'CoT-by-Design' guna membangun asisten virtual yang lebih akurat, etis, dan kredibel di lingkungan perguruan tinggi.
A DECISION SUPPORT SYSTEM MODEL FOR SELECTING SCHOLARSHIP RECIPIENTS USING A COMBINATION OF THE AHP AND TOPSIS METHODS Thamrin, Musdalifa
Nusantara Hasana Journal Vol. 4 No. 10 (2025): Nusantara Hasana Journal, March 2025
Publisher : Yayasan Nusantara Hasana Berdikari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59003/nhj.v4i10.2035

Abstract

The selection of scholarship recipients is a critical process in determining which students are eligible for educational assistance based on specific criteria. The selection process, which is still conducted manually, often leads to issues such as subjectivity, inaccurate assessments, and significant time consumption. This study aims to develop a Decision Support System model for scholarship recipient selection using a combination of the Analytical Hierarchy Process (AHP) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The AHP method is used to determine the priority weights of each criterion, while the TOPSIS method is used to rank scholarship candidates based on the highest preference scores. The results of the study indicate that the combination of these two methods is capable of producing a more objective, effective, and accurate selection process. The developed system is expected to assist educational institutions in determining scholarship recipients in a targeted and transparent manner.
THE APPLICATION OF THE ARAS METHOD IN EVALUATING THE TOP GRADUATING STUDENTS Thamrin, Musdalifa
Nusantara Hasana Journal Vol. 5 No. 8 (2026): Nusantara Hasana Journal, January 2026
Publisher : Yayasan Nusantara Hasana Berdikari

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59003/nhj.v5i8.2038

Abstract

This study aims to design and develop a decision support system for identifying the top graduating students using the Additive Ratio Assessment (ARAS) method at STMIK Profesional Makassar. This system was developed to facilitate the objective evaluation of students based on several criteria, namely GPA, TOEFL scores, academic achievements, and organizational involvement. The ARAS method was used because it is capable of performing an alternative ranking process based on predetermined criteria values and weights. Data collection was conducted through interviews, observations, and literature reviews. The system was built using the PHP programming language and a MySQL database. The results of the study indicate that the application of the ARAS method can assist the university in identifying the top students more effectively, quickly, and accurately based on the results of optimization calculations and alternative student rankings.
Co-Authors Felicia Watratan, Alvina Mallu, Satriawaty Moeis, Dikwan Sry Yunarti Yahya, Kurnia