Evaluation of students' innovation and entrepreneurship based on genetic neural network algorithm under sustainable development in higher education institutions

The study aims to assess the innovation and entrepreneurship ability of higher vocational students to support the sustainable development of higher education institutions. This paper investigates the key factors influencing students' innovation and entrepreneurship and establishes a comprehensive evaluation system. A combined weight model is constructed using the AHP and entropy method to determine the weight of the evaluation index. Genetic algorithms are used to optimise backpropagation neural networks to improve learning speed and reduce the risk of overfitting. The results showed that the proposed model exhibited lower mean square error and higher accuracy under different sample sizes and training set ratios. The comprehensive weight of evaluation indicators 1, 2 and 5 was higher, and the score of scheme A was the highest (0.536). GA-BP algorithm was superior to BPNN, random forest, and decision tree algorithms in performance. The paper presents a scientific and objective evaluation system for students' innovation and entrepreneurship abilities, which is beneficial for higher education institutions to cultivate students' innovation and entrepreneurship more effectively. In addition, the GA-BP model provides a new perspective for solving complex educational evaluation problems and promotes the development of educational evaluation methods.