Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal SainsTech Innovation Journal Journal of Vocational, Informatics and Computer Education
M. Dermawan Mulyodiputro
Universitas Bima Internasional MFH
Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Education

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

Found 2 Documents
Search

 1 
Implementing the SAW Method in a Web-Based Decision Support System for Study Program Selection: Development, Testing, and Evaluation at a Private University in Indonesia Ahmad Ryadussholihin Syafi’i; Asmaul Husna RS; Sirajun Nasihin; Mindi Richia Putri; M. Dermawan Mulyodiputro
Journal of Vocational, Informatics and Computer Education Vol 4, No 1 (2026): March 2026
Publisher : Academic Bright Collaboration

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.66053/voice.v4i1.414

Abstract

Purpose – This study aims to address the problem of inappropriate study program selection among prospective university students by developing a web-based decision support system (DSS) using the Simple Additive Weighting (SAW) method to generate objective and structured recommendations. Methods – The research employs a system development approach by implementing the SAW method in a web-based DSS. The system evaluates four cognitive criteria, namely verbal ability (weight 0.25), numerical ability (weight 0.30), logical reasoning (weight 0.25), and spatial ability (weight 0.20), applied to four D3 study program alternatives. Each criterion was weighted, normalized, and calculated to produce a ranking of study program alternatives. Data were collected through questionnaires and institutional academic records. Findings – A pilot test involving 40 high school students showed an 85% alignment rate, where most participants considered the recommendations appropriate to their interests. User acceptance scores ranged from 4.25 to 4.32 out of 5 across usability, interface clarity, and recommendation usefulness, indicating a positive initial reception. These results suggest the system demonstrates preliminary feasibility as a structured recommendation aid, though broader generalization requires further testing. Research Implications – The findings suggest that multi-criteria DSS can enhance the objectivity and consistency of study program recommendations and can be practically implemented in higher education admission processes. Originality – This study contextualizes the SAW method within a specific private university setting, applying it to four D3 study program alternatives using four cognitive criteria weights established through institutional input. The contribution lies in combining system implementation, functional testing, and user acceptance evaluation into a documented development framework applicable to higher education admission support contexts.
Benchmarking Model Machine Learning untuk Prediksi Data Berdasarkan Akurasi dan Error Valian Yoga Pudya Ardhana; Syahrani Lonang; Danang Tejo Kumoro; M. Dermawan Mulyodiputro
SainsTech Innovation Journal Vol. 8 No. 2 (2025): SIJ VOLUME 8 NOMOR 2 TAHUN 2025
Publisher : LPPM Universitas Qamarul Huda Badaruddin

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37824/sij.v8i2.2025.1141

Abstract

Perkembangan machine learning mendorong pemanfaatan berbagai model regresi untuk melakukan prediksi data secara akurat dan efisien. Namun, perbedaan karakteristik dataset menyebabkan kinerja setiap model bervariasi, sehingga diperlukan proses benchmarking untuk menentukan model yang paling optimal. Penelitian ini bertujuan untuk membandingkan kinerja beberapa model machine learning dalam tugas prediksi data berbasis regresi tanpa melakukan pengembangan aplikasi. Model yang dievaluasi meliputi Linear Regression, Decision Tree Regression, Random Forest Regression, Support Vector Regression, dan K-Nearest Neighbor Regression. Dataset yang digunakan merupakan dataset publik dengan variabel numerik yang telah melalui tahap praproses data, meliputi pembersihan data, normalisasi, dan pembagian data latih serta data uji. Evaluasi kinerja model dilakukan menggunakan metode K-Fold Cross Validation dengan metrik Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), dan koefisien determinasi (R²). Hasil penelitian menunjukkan bahwa Random Forest Regression memberikan kinerja terbaik dengan nilai error terendah, nilai R² tertinggi, serta stabilitas model yang baik dibandingkan model lainnya. Hasil ini menunjukkan bahwa pendekatan ensemble efektif dalam meningkatkan akurasi dan kemampuan generalisasi model pada tugas prediksi data regresi.
Co-Authors Ahmad Ryadussholihin Syafi’i Asmaul Husna RS Danang Tejo Kumoro Mindi Richia Putri Sirajunnasihin, Sirajunnasihin Syahrani Lonang Valian Yoga Pudya Ardhana