摘要
多模态学习分析为复杂环境下的学习测量与评价提供了新的视角。文章以涵盖多模态学习分析各类学术组织和学术论文的数据库为分析源,采用多种分析方法,对多模态学习分析的研究共同体、研究主题、研究领域、演进情况、研究热点等进行了分析,研究结果显示:多模态学习分析研究已初步形成多个学术群体,但各群体呈零散分布状态;学生编程建模、卷积与对称深层神经网络、传感器数据采集、人类行为识别、协作建构与自动评价等为主要研究主题;多模态感知与情绪分析、多模态数据融合与挖掘、多模态表征与对象识别是多模态学习分析的三大研究领域;演进情况为从知识模型与框架设计发展到多模态特征提取与情感分析、再发展到多模态表征学习与深度学习;研究热点包括多模态学习数据采集与系统设计、多模态学习认知与行为分析、多模态学习分析方法与工具开发。
Multimodal learning analysis provides a new perspective for learning measurement and evaluation in complex environments.Taking the database of all kinds of academic organizations and academic papers covering multimodal learning analysis as the analysis sources,this paper analyzed the research community,research topics,research fields,evolution situation and research hotspots of multimodal learning analysis,using a variety of analysis methods.The research results showed that the multimodal learning analysis research has initially formed multiple academic groups,while each group was in a scattered state.The students’program modeling,convolutional and symmetrical deep neural network,sensor data collection,human behavior recognition,collaborative construction and automatic evaluation were the main research topics.Meanwhile,the multimodal perception and emotion analysis,multimodal data fusion and mining,multimodal representation and object recognition were the three major research fields of the multimodal learning analysis.The evolution situation developed from knowldege model and framework design to multimodal feature extraction and emotion analysis,and then to multimodal representation learning and deep learning.The research hotspots included multimodal learning data collection and system design,multimodal learning cognition and behavior analysis,and multimodal learning analysis methods and tools development.
作者
牟智佳
符雅茹
MOU Zhi-jia;FU Ya-ru(Research Center of“Internet+Education”,Jiangnan University,Wuxi,Jiangsu,China 214122)
出处
《现代教育技术》
CSSCI
2021年第6期23-31,共9页
Modern Educational Technology
基金
江苏省高等教育教改研究重点项目“‘互联网+教育’背景下基于设计思维的课程设计与教学实践研究”(项目编号:2019JSJG082)
中央高校基本科研业务费专项资金资助课题:教育信息化2.0背景下信息技术与教学融合创新研究(项目编号:2019JDZD08)的阶段性研究成果。
关键词
多模态学习分析
多模态数据
人工智能
数据融合
multimodal learning analysis
multimodal data
artificial intelligence
data fusion
