Optimization of interdisciplinary competence development methods for university faculty based on artificial neural...

Area of Science:

• Higher Education Research
• Educational Technology
• Artificial Intelligence in Education

Background:

• Interdisciplinary competence is crucial for university faculty driving educational innovation and knowledge integration.
• Developing effective faculty development mechanisms requires precise identification and optimization of competence pathways.
• Existing methods may lack the predictive power for complex, nonlinear relationships in competence development.

Purpose of the Study:

• To propose and validate an Artificial Neural Network (ANN)-based modeling framework for identifying and optimizing interdisciplinary competence pathways in university faculty.
• To quantitatively model and predict the influence of various factors on interdisciplinary competence.
• To provide empirical evidence and methodological support for designing targeted faculty development programs.

Main Methods:

• Utilized a Multilayer Perceptron (MLP), a type of Artificial Neural Network, for quantitative modeling and nonlinear prediction.
• Evaluated 14 influencing factors on interdisciplinary competence.
• Compared MLP performance against One-dimensional Convolutional Neural Network (1D-CNN) and decision tree models using metrics like Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R²).

Main Results:

• The MLP model demonstrated superior predictive accuracy with a test set MSE of 0.276 (4.5% lower than 1D-CNN) and MAE reduced by 2.11%.
• MLP achieved a high goodness of fit (R² = 0.843), significantly outperforming 1D-CNN (0.96% higher) and decision trees (24.24% higher).
• Classification accuracy for "high" and "low" competence levels was 84.09% and 82.35% respectively, with an overall average accuracy of 79.34%.

Conclusions:

• Artificial Neural Networks, particularly MLP, offer a robust framework for modeling and predicting interdisciplinary competence in faculty.
• Cognitive flexibility is key for low-competence groups, while external resources and internal motivation drive higher competence levels.
• A parallel advancement strategy involving cognitive activation, curriculum integration, and institutional empowerment is recommended for systemic transformation in faculty development.