Construction of an Intelligent Teacher Assistant System Using the TPACK Framework and Machine Learning to Diagnose Work and Energy Misconceptions

Misconceptions in physics education, particularly in work and energy, present significant barriers to student understanding and achievement. Common misconceptions include misunderstandings of energy conservation principles, misapplying work-energy relationships, and confusion between potential and kinetic energy. These misconceptions are critical as they form the foundation for understanding more complex physics concepts. To address these challenges, this study introduces the Intelligent Teacher Assistant System (ITAS), which integrates the Technological Pedagogical Content Knowledge (TPACK) framework with machine learning to diagnose and address misconceptions in real-time. ITAS’s unique innovations include adaptive feedback tailored to individual learning needs, real-time diagnostics, and seamless alignment of technological tools with pedagogical strategies. System validation achieved a reliability score of 93.02% and usability score of 94.44%, based on standardized expert evaluations. Field testing with 150 students and 30 teachers demonstrated a 75% improvement in conceptual understanding, with average post-test scores increasing by 20%. These results underscore ITAS’s potential to transform physics education by addressing persistent misconceptions and fostering deeper student engagement. Future research will explore extending ITAS’s application to other subjects and refining its adaptive algorithms.