Bridging the Gap: Use Symbolic Language as a Prompt Between Human Emotion and Machine Logic*

This study introduces an innovative hybrid framework that utilizes symbolic language as a prompt to bridge human emotion with machine logic. With SHAP analysis validating feature contributions, visual differences between the traditional sentimental model and the current emotional classification, the approach advances emotion recognition. With well-defined, distinguishable feature classes, the proposed hybrid model improves transparency and interpretability in AI-driven decision-making.