A dynamic multimodal evaluation model for college aesthetic education integrating graph attention network and reinforcement learning

Key Points

• The aim is to develop and assess a dynamic evaluation model for enhancing college aesthetic education through innovative technologies.
• Developed a multimodal evaluation model combining graph attention networks and reinforcement learning.
• Applied this model to assess college aesthetic education programs in a randomized trial framework.
• Demonstrated significant improvements in student engagement and learning outcomes in aesthetic education.
• Showed a higher satisfaction rate among students using the new model compared to traditional methods.

Abstract