Research on Intelligent Recognition of Online Learning Cognitive Styles Based on Bimodal Data

• In Event: Artificial Intelligence Applications in Psychometrics: From Scale Development to Response Analysis

Abstract

This study investigates the identification of cognitive styles in online learning using bimodal data, including facial expressions and EEG signals. Six common machine learning algorithms were used to construct and validate the models. The results indicate that the Random Forest (RF) algorithm performed best, achieving the highest accuracy, recall, and F1 scores. Additionally, the study shows that models based on bimodal data outperform those using unimodal data, providing a more comprehensive assessment of cognitive styles. This model can enhance personalized instruction by adapting to individual cognitive preferences, thereby improving learning outcomes and satisfaction. Future research should expand the study context and explore advanced feature extraction techniques to further enhance model performance.

Authors

• Zhan Chen, East China Normal University

• Yaofeng Xue, East China Normal University

• Yisheng Qiu, East China Normal University