Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal bit-Tech Surya Informatika
Pradana, Novant Nanda
Unknown Affiliation
Computer Science & IT

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

Found 2 Documents
Search

 1 
Implementation of an Employee Attendance System with Web and Mobile-Based Face Recognition Technology (OpenCV) Pradana, Novant Nanda; Sardi, Sophian Andhika; Alfin, Anggi
bit-Tech Vol. 8 No. 3 (2026): bit-Tech
Publisher : Komunitas Dosen Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32877/bt.v8i3.3737

Abstract

The attendance system implemented at PT Sinergi Karya Mandiri previously depended on a paid third-party platform that still presented several weaknesses in validation accuracy. The platform only allowed manual photo uploads, creating opportunities for attendance manipulation and making employee location verification difficult to ensure. This study was conducted to design and implement an attendance system based on facial recognition technology using OpenCV in order to improve the reliability, validity, and transparency of attendance records. The proposed solution was developed as an integrated system consisting of a mobile application for employees to perform attendance through face verification and GPS-based location capture, as well as a web-based application that enables the HRD division to monitor attendance activities and generate reports in real time. System development adopted the Rapid Application Development (RAD) approach, which covers the stages of Requirements Planning, User Design, Construction, and Cutover. The implementation results indicate that the system was able to apply facial recognition using the Local Binary Pattern Histogram (LBPH) method, with the recognition process executed locally on the mobile device. This mechanism supports offline functionality while also providing faster response times. Based on black box testing, all major features operated properly and produced valid results, including HRD login, employee OTP verification, facial-recognition-based attendance, employee data management, attendance location mapping, and offline synchronization reporting. Overall, the developed system reduced dependence on external paid services, minimized fraud potential, and improved operational efficiency in attendance management.
Perbandingan Kinerja Algoritma Machine Learning Dalam Prediksi Kesehatan Mental Dan Burnout Mahasiswa Hidayat, Jose Julian; Azhari, Fajri Fauzan; Husna, Tsania Manzilatul; Fahmayani , Aulia Nufaila; Pradana, Novant Nanda; Setyowati, Cindy
Jurnal Surya Informatika Vol. 16 No. 1 (2026): Jurnal Surya Informatika, Vol 16. No. 1, Mei 2026
Publisher : Universitas Muhammadiyah Pekajangan Pekalongan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.48144/suryainformatika.v16i1.2420

Abstract

Penelitian ini bertujuan untuk membandingkan kinerja beberapa algoritma machine learning dalam memprediksi tingkat kesehatan mental dan burnout mahasiswa, yang diklasifikasikan ke dalam tiga kategori, yaitu Low, Medium, dan High. Algoritma yang diuji meliputi Decision Tree, Logistic Regression, Naive Bayes, Support Vector Machine (SVM), dan Random Forest. Evaluasi dilakukan menggunakan metrik Accuracy, Precision, Recall, dan F1-Score pada dataset berjumlah 200.000 data. Hasil penelitian menunjukkan bahwa Logistic Regression memiliki performa terbaik secara keseluruhan dengan nilai akurasi sebesar 0,8720 dan F1-Score sebesar 0,8677, diikuti oleh Random Forest dengan akurasi 0,8708. Decision Tree dan SVM juga menunjukkan performa yang kompetitif dengan akurasi masing-masing sebesar 0,8646 dan 0,8684, sementara Naive Bayes memiliki performa terendah dengan akurasi 0,8503. Namun demikian, seluruh model mengalami kesulitan dalam memprediksi kelas High, yang ditunjukkan oleh nilai recall yang relatif rendah, terutama pada SVM yang gagal mendeteksi kelas tersebut. Hal ini mengindikasikan adanya ketidakseimbangan data yang signifikan, di mana kelas Low mendominasi dataset. Secara keseluruhan, Logistic Regression dan Random Forest dapat direkomendasikan sebagai model terbaik untuk prediksi kesehatan mental mahasiswa dalam studi ini. Namun, diperlukan strategi penanganan data tidak seimbang, seperti resampling atau cost-sensitive learning, untuk meningkatkan performa prediksi pada kelas minoritas, khususnya kategori High.
Co-Authors Alfin, Anggi Azhari, Fajri Fauzan Fahmayani , Aulia Nufaila Hidayat, Jose Julian Husna, Tsania Manzilatul Sardi, Sophian Andhika Setyowati, Cindy