Computational framework for ideological and political teaching of intelligent courses: adaptive recommendation driven by integrating multi-modal learning analysis, emotion recognition and generative AI

This research is oriented to ideological and political teaching of university courses, and construct an intelligent teaching computing framework that integrates multi-modal learning analysis, emotion recognition and generative AI recommendation. By collect students' voice, expression and behavior data in class, the system identifies the learning state and emotional response in real time, and pushes personalized teaching resources accordingly. The experiment was carried out in a total of 460 student in many universities, designed as a control group, and the system covered the whole teaching process. The results showed that the students in the experimental group were better than the control group in terms of learning behavior activity, emotional participation, tests scores and satisfaction. Further analysis showed that the frequency of recommended resource use, behavioral engagement and emotional state are keys factors influencing learning effectiveness. The research verifies the application potential of artificial intelligence technology in improving quality of ideological and political teaching, and provide a practical basis for teach system design.