Comprehensive Feature Analysis and Evaluation on the Student Performance Based on Machine Learning

The cultivation of high-quality talents relies on the synergistic interaction of various educational stakeholders, including schools, families, and society, within the educational system. With the rapid advancement of artificial intelligence (AI) technology, new opportunities have emerged for constructing and optimizing collaborative education mechanisms. Based on a feature-rich and large-scale real dataset, this paper conducts a case study to explore novel approaches for leveraging AI to empower a cooperative education system. Specifically, correlation and association analysis methods from traditional statistics are first employed to quantify pairwise feature relationships, providing a basis for identifying key factors influencing student development. Subsequently, principal component analysis (PCA) is applied to extract dominant components from the dataset, assess the intrinsic information carried by each feature, and uncover latent relationships among features. Finally, leveraging the multi-source and heterogeneous nature of the cooperative education system, a novel multi-branch neural network model (MBDNN) is proposed to achieve accurate prediction of student academic performance. This study can provide reference and methodological support for effectiveness evaluation and decision-making within the cooperative education system.