Research on the evaluation of the effect of moraleducation integration in college students physicaleducation classroom based on behavioral perceptionmodel

In the rapidly evolving landscape of computational sciences, the integration of moral education within physical education(PE) classes offers an innovative and impactful avenue for fostering ethical development among college students. Byintroducing a computational methodology to assess moral education in PE contexts, this study transcends traditionalevaluation techniques, which often rely on qualitative and anecdotal assessments lacking scalability, objectivity, andpredictive capabilities. To mitigate the aforementioned shortcomings, we design a novel architecture built upon two coremechanisms—Epistemic Trajectory Encoder (ETE) and Knowledge-Constrained Adaptive Inference (KCAI). The ETE isdesigned to capture the temporal progression of students’ moral development by modeling their ethical decision-makingover time, while the KCAI module introduces domain-specific knowledge constraints to refine the inference process,ensuring both contextual relevance and higher interpretability. This dual-component system enables the detailed modelingof students’ behavioral responses and ethical reasoning in PE environments, allowing educators and researchers toderive actionable insights. Empirical results from our experiments indicate that the proposed model significantly outperforms existing benchmarks in predicting and understanding moral behavior trajectories. These findings underscore thetransformative potential of computational intelligence in educational environments, not only enhancing the accuracy ofassessments but also contributing to the broader discourse on how technology can support character education. Ultimately,this approach serves as a robust tool for promoting ethical awareness through enriched, data-driven physical education curricula.