The evaluation model of engineering practice teaching with complex network analytic hierarchy process based on deep...

Area of Science:

• Engineering Education
• Computer Science
• Data Science

Background:

• Traditional engineering practice teaching evaluation systems often lack efficiency and comprehensive management.
• There is a need for advanced technological solutions to improve the quality and adaptability of educational assessment.

Purpose of the Study:

• To develop and validate a novel teaching evaluation model for engineering practice in universities.
• To enhance the efficiency of teaching management and provide data-driven insights into student learning.

Main Methods:

• Utilized Internet of Things (IoT) technology, complex network analytic hierarchy process, and deep learning (RNN, CNN).
• Employed Natural Language Processing (NLP) and Generative Adversarial Network (GAN) for feature extraction from online course data.
• Integrated real curriculum data with simulated student data for model training and analysis.

Main Results:

• A positive correlation (r=0.725) was found between curriculum dependence and interdisciplinary impact.
• The deep learning model achieved 76-98% accuracy and prediction consistency in evaluating student performance.
• Identified key factors influencing learning outcomes and revealed trends in student learning processes.

Conclusions:

• The developed model effectively improves engineering practice teaching evaluation and management efficiency.
• The study provides valuable insights into the relationship between teaching factors and student performance.
• The dynamic and adaptive nature of the model allows for continuous improvement in educational environments.