中国教育类核心期刊 CSSCI来源期刊 RCCSE中国权威学术期刊

精品文章

您的位置:首页 >精品文章>

    多模态学习情感计算:动因、框架与建议

    Multimodal Learning Affective Computing: Motivations, Frameworks and Suggestions

【作      者】:

周 进, 叶俊民, 李 超


【关 键 词 】:

情感计算; 多模态; 人工智能; 学习情感; 学习分析


【栏      目】:

理论探讨


【中文摘要】:

学习情感是影响学生认知加工与学习效果的重要因素,如何利用多模态数据开展学习情感计算是当前亟待解决的问题。文章在分析情感计算源起与多模态数据融合的基础上,阐述了多模态情感计算的发展动因,构建了多模态学习情感计算的研究框架,包括以教育场景为导向采集情感数据、依据情感模型展开建模与识别、利用可视化方式表达与反馈情感、结合情感归因来干预与调节学习过程等。基于现有研究案例,将多模态学习情感计算的应用归纳为开发学习情感识别系统、增强智能学习工具、支持学习干预与决策、探索学习情感的作用机制等方面。未来多模态学习情感计算应平衡数据采集侵入性与真实性、提升数据模型可解释性、综合衡量学习状态以及拓展教育应用探索与创新。


【英文摘要】:

Learning emotion is an important factor affecting students' cognitive processing and learning outcomes, and how to use multimodal data to carry out learning affective computing is an urgent problem to be solved. Based on the analysis of the origin of affective computing and the fusion of multimodal data, this paper expounds the motivation for the development of multimodal affective computing and constructs a research framework for Multimodal Learning Affective Computing (MLAC), including collecting affective data based on educational scenarios, modeling and identifying based on affective models, expressing and giving feedback on emotions using visualization, and combining affective attribution to intervene and regulate the learning process. Based on existing research cases, the applications of MLAC are summarized as developing learning emotion recognition systems, enhancing intelligent learning tools, supporting learning interventions and decision-making, and exploring the mechanisms of learning emotion. In the future, MLAC should balance the intrusiveness and authenticity of data collection, enhance the interpretability of data models, comprehensively measure learning states, and expand the exploration and innovation of educational applications.


【下      载】:

多模态学习情感计算:动因、框架与建议

网站首页 | 联系我们 | 意见建议| 杂志订阅| 管理部登录

版权所有 © 电化教育研究 CopyRight ©e-EDUCATION RESEARCH 2011-2018 电话:0931-7971823 地址:西北师范大学《电化教育研究》杂志社