Design and Development of Sensor Driven Dynamic Time Warping Model for Digital Online Art Teaching System Based

Traditional art education often relies on static teaching methods and materials, which may not effectively engage students or cater to diverse learning styles. These conventional approaches can lead to decreased student motivation and limited opportunities for interactive learning experiences. To address these challenges, this study proposes a digital online art teaching system that leverages the Sensor driven Dynamic Time Warping(SD-DWT) algorithm to enhance interactivity and personalize learning. DTW is adept at measuring similarities between time-series data, making it suitable for aligning and comparing students' artistic progress over time. The system utilizes the Art Dataset for Interactive System and Sensor Data, which comprises sensor-based interaction logs and user behavior data collected during digital art activities. By applying SD-DTW to analyze students' interaction patterns, the system can adaptively provide feedback and tailor instructional content to individual learning trajectories. Experimental results demonstrate that the SD-DTW-based system achieves an accuracy of 97.8%, precision of 96.5%, recall of 96.9%, F1-score of 96.7%, specificity of 98.1%, Matthews Correlation Coefficient (MCC) of 0.951, and an AUC-ROC score of 0.982 in predicting and enhancing student engagement levels. These findings underscore the potential of integrating DTW algorithms into digital art education platforms to foster personalized and effective learning experiences.