A Student Education Data Mining Method based on Student Sequential Behaviors and Hybrid Recurrent Neural Network

Abstract

This research aims to propose a student education data mining method based on a hybrid recurrent neural network and improved support vector machine-decision tree algorithm through in-depth analysis of student behavior sequences. The method combines the feature extraction capability of a hybrid recurrent neural network and the nonlinear mining efficiency of support vector machine-decision tree algorithm to achieve efficient prediction of students’ learning behavior and performance. Experimental results have shown that the designed method completed an APA of 91.3% and 89.2% for the HR-SDT model on the Student_1 and Student_2 datasets, respectively. The F1 score average values of the HR-SDT model reached 86.7% and 81.9%, respectively. The results indicate that the student behavior data mining method based on hybrid recurrent neural networks can accurately predict the learning behavior and performance of students, providing valuable insights and decision support for educators.