Article Per Year (5 Year)
p-Index From 2021 - 2026
0.444
P-Index
This Author published in this journals
All Journal Jurnal Sistem Komputer dan Informatika (JSON) Jurnal Penjaminan Mutu
Widya Dharma Sidi
Unknown Affiliation
Computer Science & IT Control & Systems Engineering Education Other

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

Found 2 Documents
Search

 1 
EVALUASI USABILITY SIKARSA (SISTEM INFORMASI KEARSIPAN UHNSUGRIWA) DAN ANALISIS KORELASI KOEFISIEN USABILITY DENGAN MODEL INSTRUMEN SYSTEM USABILITY SCALE Dewa, I Dewa Gede Budiastawa; Widya Dharma Sidi; I Made Agus Oka Gunawan; Gede Indrawan
Jurnal Penjaminan Mutu Vol 11 No 01 (2025)
Publisher : UHN IGB Sugriwa Denpasar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25078/jpm.v11i01.4408

Abstract

I Gusti Bagus Sugriwa Denpasar Hindu State University is one of the Hindu State Religious Colleges which is directly under the Ministry of Religion. The many bureaucratic processes result in the production of many documents ranging from planning documents to reporting documents. In the midst of planning and reporting, there are various processes that will certainly produce various documents. The documents produced will be needed at any time, starting from the accreditation process to the accreditation institution to the audit process carried out by the inspectorate general. Ease of access to documents requires an effective and efficient archiving method, so the Uhnsugriwa Archiving Information System (SIKARSA) was developed. Since it was developed in 2021 and used effectively in 2022, this archiving information system has not been tested, especially at the level of usability, so testing is needed to determine the level of usability of the information system and provide recommendations for strategic steps in the future to maximize the function of the archiving information system. Based on the results of the analysis that has been carried out using the System Usability Scale (SUS) model and target respondents from 4 (four) types of users, namely echelon I - IV officials, non-structural officials, archivist functional officials, and other functional officials / lecturers / implementers, the results of the evaluation of the usability of the archiving information system are in the very good category with details of the category of non-structural official users of 90.75, followed by echelon I - IV officials with a value of 85.625, followed by archivist functional officials who are the main figures in archival application users of 81.25, and the category of other functional officials / lecturers / implementers with a value of 76.17, the results of the analysis show that users give values ​​in the range of 75 - 100, so that the information system is included in the appropriate and acceptable category.
Dominansi Fitur Perilaku Keuangan dan Tren Akademik pada Deteksi Dini Kelulusan Mahasiswa dengan XGBoost Widya Dharma Sidi
Jurnal Sistem Komputer dan Informatika (JSON) Vol. 7 No. 3 (2026): Maret 2026
Publisher : Universitas Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/json.v7i3.9591

Abstract

Ketepatan waktu kelulusan merupakan indikator krusial bagi keberlanjutan institusi pendidikan tinggi. Namun, metode pemantauan konvensional sering kali luput mendeteksi mahasiswa berisiko secara dini karena hanya mengandalkan atribut demografi statis dan mengabaikan dinamika perilaku mahasiswa. Penelitian ini bertujuan mengembangkan model deteksi dini keterlambatan lulus yang berfokus pada fitur perilaku keuangan dan tren akademik pada dataset berskala kecil dan tidak seimbang di Universitas PGRI Mahadewa Indonesia menggunakan algoritma Extreme Gradient Boosting (XGBoost). Tantangan berupa ketidakseimbangan data (imbalanced data) dengan baseline accuracy sebesar 67,61% diatasi melalui teknik Synthetic Minority Over-sampling Technique (SMOTE). Hasil eksperimen menunjukkan performa model yang andal dengan rata-rata F1-Score sebesar 81,56% dan Recall 80,81%, yang membuktikan efektivitas sistem sebagai solusi dalam meminimalkan kesalahan deteksi pada mahasiswa berisiko. Analisis Feature Importance mengungkapkan bahwa variabel perilaku dinamis lebih dominan dibandingkan faktor demografi statis. IPS Semester 4 (Gain 0,269) dan Frekuensi Tunggakan (Gain 0,151) ditemukan sebagai prediktor paling signifikan. Temuan ini mengonfirmasi bahwa performa akademik di fase menengah dan kedisiplinan administrasi keuangan merupakan indikator kritis kelulusan. Penelitian ini merekomendasikan penerapan Sistem Deteksi Hibrida pada akhir semester 4 untuk memberikan intervensi tepat sasaran bagi mahasiswa di Zona Merah.
Co-Authors Dewa, I Dewa Gede Budiastawa Gede Indrawan I Made Agus Oka Gunawan