基于多模态数据的在线学习认知风格智能识别被引量：4

Intelligent Recognition of Cognitive Styles in Online Learning Based on Multimodal Data

在线阅读下载PDF

导出

摘要在人工智能技术迅速发展的背景下,多模态数据分析在教育领域的重要性日益凸显。本研究通过提取学习者在线学习中产生的脑电数据、表情数据和眼动数据等,构建了基于多模态数据的认知风格智能识别框架,并运用六种机器学习模型验证其有效性。研究结果显示,四种机器学习模型的识别准确率超过80%,且最佳模型的准确率达89.17%,F1分数达0.9241。此外,与基于单模态数据的模型相比,基于多模态数据的认知风格模型识别表现更佳。这表明,多模态数据融合策略具有优越性,有助于加强在线学习平台的适应性和个性化。 With the rapid advancement of artificial intelligence technology,the importance of multimodal data analysis in education is becoming increasingly critical.This study collected various data types during learners'online learning processes,including electroencephalogram(EEG),facial expression,and eye-tracking data,to construct an intelligent cognitive style recognition framework through feature extraction and fusion processing.The effectiveness of this framework was validated using six machine-learning models.The optimal model was identified through comparison.The results showed that four machine learning models achieved recognition accuracy rates exceeding 80%,with the best model reaching an accuracy of 89.17%with an F1 score of 0.9241,demonstrating high recognition precision.Furthermore,the study compared models based on single-modal data and multimodal data and found that the model on multimodal data performed significantly better in cognitive style recognition,providing evidence for the multimodal data fusion strategy,which contributes to enhancing the adaptability and personalization of online learning platforms.

作者薛耀锋陈瞻邱奕盛朱芳清 XUE Yaofeng;CHEN Zhan;QIU Yisheng;ZHU Fangqing(Department of Educational Information Technology&Shanghai Engineering Research Center of Digital Education Equipment,East China Normal University,Shanghai 200062,China)

机构地区华东师范大学教育信息技术学系

出处《开放教育研究》 CSSCI 北大核心 2024年第5期112-120,共9页 Open Education Research

基金 2022年度上海市自然科学基金项目“基于多模态数据融合的在线学习认知模型及优化研究”(22ZR1421300)。

关键词多模态数据在线学习认知风格脑电信号面部表情眼动追踪 multimodal data online learning cognitive styles electroencephalogram(EEG)signals facial expressions eye-tracking

分类号 G434 [文化科学—教育技术学]

作者简介薛耀锋,博士,副研究员,硕士生导师,华东师范大学教育信息技术学系,研究方向:智能教育技术(yaofeng.xue@163.com);陈瞻,硕士研究生,华东师范大学教育信息技术学系,研究方向:智能教育技术;邱奕盛,硕士研究生,华东师范大学教育信息技术学系,研究方向:智能教育技术;朱芳清,硕士研究生,华东师范大学教育信息技术学系,研究方向:智能教育技术。