Research on the Construction of Higher Education Ecosystem Model under the Empowerment Development Strategy of Higher Education

Abstract

The construction of a higher education ecosystem model is a crucial research direction for enhancing the quality and management efficiency of higher education. Based on the empowerment development strategy of higher education, this study adopts graph neural networks (GNNs) combined with genetic algorithms (GAs) for system modeling and optimization. A complex dynamic system model encompassing multiple elements such as teachers, students, courses, and teaching resources is constructed. Experimental results demonstrate that GA-GNN exhibits outstanding performance across multiple key indicators. In the robustness analysis, the performance score of GA-GNN gradually increases from an initial 0.509 to 0.891 after 800 iterations, showcasing a consistent and stable improvement trend. In terms of generalization ability, GA-GNN achieves a score of 0.895 on the university performance dataset and 0.942 in the direction of curriculum design improvement, indicating its broad adaptability across different datasets and application scenarios. Furthermore, GA-GNN also performs exceptionally well in convergence, with an initial convergence score of 0.440 and reaching 0.950 after 800 iterations, far surpassing the performance of other algorithms.In summary, GA-GNN demonstrates wide applicability and excellent performance in higher education management, possessing efficient modeling capabilities and application value within the complex and dynamic higher education ecosystem.