Research on the Integration of Student Behavior Analysis and Curriculum Education Strategies in Colleges and Universities under Deep Learning Framework

This paper designs a student behavior analysis system for colleges and universities based on a deep learning framework. The system can collect and analyze students’ behaviors, and teachers can provide personalized teaching guidance based on the analysis results, thus assisting in curriculum education reform. The system’s behavior analysis module is designed from the perspective of human skeletal joint points. The Gaussian noise in the image is first eliminated by Gaussian filtering. Then the position of the target student in the image is detected by the target detection algorithm that incorporates the attention mechanism. The coordinates of human skeletal joints are extracted from the detected image using the improved OpenPose model. Finally, a support vector machine is used to classify the acquired joint coordinates to quickly and accurately identify the behavioral state of the student. The method achieves 98.04% accuracy in behavior recognition, which is 5.89% higher than CNN-10. Students’ evaluation of the educational effect of the course integrating student behavior analysis is significantly higher than that of traditional teaching. The method designed in this paper for student behavior analysis in colleges and universities, applied to course education, can improve student learning.