<![CDATA[Teaching computational physics and developing programming skills remains a significant challenge for many undergraduate programs worldwide. This study presents an innovative approach implemented at the University of Mataram, Indonesia, to optimize Pascal programming instruction for physics students. Various artificial intelligence (AI) techniques were utilized to assist students in developing more complex programs. This enabled the generation of customized lesson plans with interactive tutorials, coding exercises, and simulations tailored to each student's needs. Throughout the semester, the system continuously monitored student progress and adjusted instructional materials accordingly. The analysis of Mann-Whitney test results for Computational Physics scores among Classes A, B, and C revealed no statistically significant differences between the groups, with median scores consistently at 80 and p-values exceeding 0.05. However, an examination of the theoretical section of the final examination showed an overall improvement in average scores compared to the previous year, with Class A achieving the highest mean score of 84%. Additionally, the practical programming section demonstrated increased pass rates across all three classes, with Class B achieving the highest pass rate at 92%, followed by Class A at 88% and Class C at 82%. Student feedback indicated high levels of satisfaction with the approach, citing increased engagement and motivation. The study highlights the potential of leveraging AI techniques in generating personalized programming examples for computational physics education, enhancing comprehension of theoretical concepts, and facilitating the development of practical programming skills.]]>
Copyrights © 2024
