Research on Academic and Career Planning Model Based on Graph Neural Network and Transfer Learning

This paper presents an academic and career planning model based on Graph Neural Networks (GNN) and Transfer Learning, aimed at addressing the challenges of major selection and career planning in the context of college entrance exam recommendations. The proposed model utilizes GNN to model the relationships between students' subject choices and career inclinations, effectively capturing complex relational information. To mitigate the challenges posed by few-shot dataset, Transfer Learning is introduced by incorporating pre-trained models, which enhance the generalization capability of the GNN. Additionally, the SMOTE algorithm is employed to handle the class imbalance in the dataset, further improving the classification performance of the model. Experimental results demonstrate that the proposed model performs exceptionally well on imbalanced datasets, providing more scientifically sound and personalized recommendations for students' major selection. The model offers enhanced guidance for students' academic and career planning, ensuring more accurate and tailored advice.