An Intelligent Assessment of College Students' Innovation and Entrepreneurship Ability Based on Optimized Backpropagation (BP) Neural Network with BAT

Résumé fourni par la source

Evaluating the innovation and entrepreneurship abilities of college students is essential for enhancing talent cultivation and guiding educational reforms. This study proposes an intelligent assessment framework based on the Backpropagation (BP) Neural Network and BAT Optimization Model, designed to quantitatively evaluate students' competencies across multiple dimensions, including creativity, opportunity recognition, risk tolerance, resource integration, and execution ability. A structured evaluation index system was developed through expert consultation and Delphi method, and survey data were collected from over$\mathbf{1, 2 0 0}$college students across various disciplines. The BP neural network model was trained on 80% of the dataset and tested on the remaining 20%, achieving a prediction accuracy of 92.3%, with a Mean Squared Error (MSE) of$\mathbf{0. 0 1 6}$. Comparative analysis with traditional statistical models (e.g., logistic regression) highlighted the superior nonlinear learning and generalization capability of the BAT optimized BP model. The proposed system not only provides an objective method for student capability assessment but also offers valuable insights for designing personalized training programs and innovation-focused curricula.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.