Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal Informatics and Computer Engineering Journal Nuansa Informatika
Sardiarinto
Universitas Bina Sarana Informatika
Computer Science & IT

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

Found 2 Documents
Search

 1 
Implementation Of Extreme Programming Method In Financial Reporting Application With Laravel Framework Akhmad Syukron; Andria Bas Nando; Sardiarinto; Eko Saputro
NUANSA INFORMATIKA Vol. 20 No. 2 (2026): Nuansa Informatika 20.2 July 2026
Publisher : FKOM UNIKU

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25134/ilkom.v20i2.588

Abstract

The advancement of information technology compels companies to adopt more efficient, accurate, and integrated approaches to financial data management. CV Mawar Magenta, a service-oriented company, still relies on manual bookkeeping and separate Microsoft Excel files for financial reporting. This method leads to duplicate tasks, a high risk of data entry errors, and delays in generating financial reports. To address these issues, a web-based Financial Reporting Application was designed and developed using the Laravel Framework. The system development employed the Extreme Programming (XP) methodology and includes core features such as cash inflow and outflow records, general journal, ledger, profit and loss reports, and balance sheet. The system was tested using the black-box testing method to ensure all functionalities align with user requirements. The implementation results show that the application simplifies financial record-keeping by integrating previously separate processes into a single system. Additionally, it streamlines data retrieval and enables business owners to monitor financial conditions in real time, thereby supporting faster and more accurate decision-making.
Analisis Perbandingan Algoritma Semi-Supervised Learning dalam Klasifikasi Adiksi Smartphone dengan Keterbatasan Data Terlabel Triadi Widianto; Nani Purwati; Hidayat Muhammad Nur; Ahmad Syukron; Sardiarinto; Eko Saputro
Informatics and Computer Engineering Journal Vol 6 No 2 (2026): Periode Juli 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat (LPPM) Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/icej.v6i2.12939

Abstract

Penelitian ini menganalisis perbandingan kinerja algoritma Semi-Supervised Learning (SSL), Support Vector Machine (SVM), dan Neural Network dalam klasifikasi adiksi smartphone pada kondisi keterbata- san data terlabel. Dataset yang digunakan berjumlah 7.500 sampel dengan 15 variabel perilaku digital dan gaya hidup. Skenario eksperimen disusun secara realistis dengan hanya menggunakan 2% data latih berlabel dan 98% data tidak berlabel untuk merepresentasikan fenomena label scarcity. Hasil pengujian menunjukkan bahwa SSL mencapai performa terbaik dengan akurasi sebesar 88,09%, melampaui SVM (86,13%) dan Neural Network (85,96%). Analisis confusion matrix mengonfirmasi bahwa SSL mem- berikan prediksi yang lebih seimbang antarkelas serta mampu menurunkan kesalahan prediksi kritis dibanding model supervised murni. Meskipun demikian, terdapat trade-off di mana SSL membutuhkan biaya komputasi yang jauh lebih tinggi, dengan waktu pelatihan mencapai 21,76 detik berbanding SVM yang hanya 0,006 detik. Temuan ini menegaskan bahwa pemanfaatan data tidak berlabel melalui mekanisme pseudo-labeling merupakan strategi efektif untuk meningkatkan kualitas deteksi dini adiksi smartphone ketika proses pelabelan data oleh ahli bersifat mahal dan terbatas.
Co-Authors Ahmad Syukron Akhmad Syukron Andria Bas Nando Eko Saputro Hidayat Muhammad Nur Nani Purwati Triadi Widianto