Technical Evaluation of the Impact of Industry-Education Integration on College Students' Employment Rate Based on Machine Learning Models

This study explores the impact of industry-education integration on college students' employment rate using machine learning models. The original data was preprocessed through feature engineering, and models such as logistic regression, decision tree, and random forest were used, with hyperparameters optimized using techniques like grid search. The experimental results show that, among 58,342 graduate data, the neural network model performs best in evaluation metrics such as accuracy, recall, precision, and F1 score. The model is most accurate in predicting medium employment quality, with errors stemming from feature noise, nonlinear relationships, and data imbalance. Analysis based on major categories and school types indicates that the model performs better on data from science and engineering disciplines and key universities. This study contributes to the formulation of relevant educational policies.