Evaluation method of college students’ education based on artificial neural network

This study constructs an educational evaluation model based on a convolutional neural network (CNN) to address the subjectivity and single-indicator limitations of traditional college student evaluation methods. Using questionnaire data collected across multiple universities in eastern, central, and western China, the model integrates multidimensional indicators including academic performance, learning process, learning attitude, and comprehensive quality. After systematic parameter optimization through grid search and Bayesian methods, the CNN model achieves 90.2% accuracy, 0.89 precision, 0.87 recall, and 0.88 F1-score on the test set. Comparative experiments show that the proposed CNN outperforms DNN (F1 = 0.86) and Random Forest (F1 = 0.83) in handling nonlinear relationships among educational indicators. The results demonstrate that the neural network-based approach enhances evaluation objectivity and provides data-driven support for teaching improvement in higher education.