<![CDATA[Prolonged sitting during learning activities is one of the factors contributing to an increased risk of musculoskeletal disorders, particularly Low Back Pain (LBP), among university students. Therefore, this study proposes an Internet of Things (IoT)-based monitoring system capable of recognizing users' activities while automatically providing rest reminders. The proposed system integrates two MPU6050 sensors, one BME280 sensor, and the Extreme Gradient Boosting (XGBoost) algorithm for activity classification. The research methodology consists of data acquisition from 25 participants, data preprocessing, feature extraction and selection, model training, evaluation using the Leave-One-Subject-Out (LOSO) cross-validation method, and deployment of the trained model in a web-based monitoring application. Experimental results demonstrate that the XGBoost model achieves an average classification accuracy of 89.11% in distinguishing sitting and standing activities. Furthermore, the developed system is capable of monitoring sitting duration in real time and automatically generating a microbreak reminder after users maintain a sitting posture continuously for 30 minutes. These findings indicate that the integration of wearable sensors, IoT, and XGBoost has the potential to support continuous monitoring of students' learning activities while helping reduce sedentary behavior among university students.]]>
Copyrights © 2026
