<![CDATA[ASD is a neurodevelopmental disorder that affects a child's ability to regulate emotions, interact socially, and respond to environmental stimuli. Monitoring physiological conditions in children with ASD is challenging because it relies on subjective observations. In this study, physiological conditions are defined based on observable behavioral states, namely active and quiet. The active state reflects increased physiological arousal and a higher heart rate, while the quiet state represents a resting state with a more stable heart rate. This study proposes an ECG-based classification system to distinguish between these two states. The dataset consists of 2000 samples for each class. Due to noise in the ECG signal caused by body movements, preprocessing was performed using the DWT to improve signal quality. The processed signals were then classified using three machine learning algorithms: SVM, Random Forest, and AdaBoost. The performance of each model was evaluated using accuracy, precision, recall, and F1-score. The results showed that without DWT, Random Forest achieved the highest accuracy of 91.00%, followed by SVM at 88.87%, and AdaBoost at 87.25%. While using DWT, Random Forest achieved the highest accuracy of 93.75%, followed by SVM at 91.37%, and AdaBoost at 90.25%. This indicates that DWT can produce better signal quality. Furthermore, Random Forest was selected as the optimal model and implemented in a Streamlit-based web application for real-time monitoring. These findings indicate that the combination of DWT and Random Forest is effective for classifying physiological conditions in children with ASD and has potential as an objective monitoring tool.]]>
