DEVELOPMENT OF AN INTELLIGENT ATTENDANCE CLOCKING SYSTEM LEVERAGING MACHINE LEARNING USING ADAPTIVE BOOSTING (ADABOOST) TECHNIQUES

Abstract

The growing demand for accuracy and automation in attendance monitoring systems, especially in educational and 
corporate institutions, has driven the need for intelligent solutions that can minimize human errors, detect anomalies, 
and ensure timely record-keeping. This study presents the development and implementation of an intelligent 
attendance clocking system using the Adaptive Boosting (AdaBoost) algorithm which is a powerful ensemble 
machine learning technique. The system processes a range of features including timestamp logs, user identifiers, and 
historical attendance behavior to classify attendance status into categories such as Present, Late, or Absent. The 
resulting system offers a scalable, intelligent, and automated framework for real-time attendance tracking, capable of 
reducing manual intervention and enhancing data integrity. Its success highlights the effectiveness of ensemble 
learning, specifically AdaBoost, in classification tasks within time-sensitive administrative domains. This approach 
can be extended to support anomaly detection, fraud prevention, and predictive scheduling in future iterations. The 
model was trained and evaluated using real-world datasets, achieving improved accuracy compared to traditional 
methods. The implementation supports real-time clock-ins, secure data handling, and efficient user management. 
The outcome demonstrates that machine learning, particularly AdaBoost, can significantly optimize attendance 
systems, making them more intelligent, scalable, and resilient to manipulation