Implementation of Face Recognition for Lecturer Attendance Using Deep Learning CNN Algorithm

Fajhar Muhammad (Informatics Study Program, Faculty of Communication and Information Technology, Universitas Nasional, Jakarta, Indonesia)
Agung Triayudi (Informatics Study Program, Faculty of Communication and Information Technology, Universitas Nasional, Jakarta, Indonesia)
Eri Mardiani (Informatics Study Program, Faculty of Communication and Information Technology, Universitas Nasional, Jakarta, Indonesia)

Journal of Blockchain, Nfts and Metaverse Technology
Vol. 2 No. 1 (2024): February 2024

Using the Convolutional Neural Network (CNN) algorithm, this research aims to create a better lecturer attendance application that improves the attendance system and creates peace of mind when lecturers arrive at national universities. The author analyses the results of applying deep learning algorithms to an experimental face recognition system that uses convolutional neural networks. The purpose of this study is to show that deep learning algorithms can improve the accuracy and efficiency of recording presence. In addition, the goal of this research is to create a timekeeping application using face recognition technology that is expected to have a high level of accuracy. In addition, this research includes a modification of the CNN model. This modification resulted in an epoch value of 75 for training of 100% and test of 95%. Analysis of results, drawing conclusions, and suggestions for additional development are the final stages of this research. Evaluation of the integrated system is done by collecting actual attendance data and comparing it with the attendance records created by the system. This validation will help explain the performance of the system and find problems or vulnerabilities that may need to be fixed.