[1]王志锋,熊莎莎,左明章,等.智慧教育视域下的知识追踪:现状、框架及趋势[J].远程教育杂志,2021,(05):045-54.[doi:10.15881/j.cnki.cn33-1304/g4.2021.05.005]
 Wang Zhifeng,Xiong Shasha,Zuo Mingzhang,et al.Knowledge Tracing from the Perspective of Smart Education: Current Situation, Framework and Trend[J].Distance Education Journal,2021,(05):045-54.[doi:10.15881/j.cnki.cn33-1304/g4.2021.05.005]
点击复制

智慧教育视域下的知识追踪:现状、框架及趋势()
分享到:

《远程教育杂志》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2021年05期
页码:
045-54
栏目:
前沿探索
出版日期:
2021-09-25

文章信息/Info

Title:
Knowledge Tracing from the Perspective of Smart Education: Current Situation, Framework and Trend
作者:
王志锋; 熊莎莎; 左明章; 闵秋莎; 叶俊民
1.华中师范大学 人工智能教育学部 教育信息技术学院;2. 华中师范大学 计算机学院,湖北武汉 430079
Author(s):
Wang Zhifeng; Xiong Shasha; Zuo Mingzhang; Min Qiusha; Ye Junmin
1.Faculty of Artificial Intelligence in Education, School of Educational Information Technology, Central China Normal University; 2. School of Computer, Central China Normal University, Wuhan Hubei 430079
关键词:
知识追踪 智慧教育 个性化教育资源推荐深度学习
Keywords:
Knowledge Tracing Smart Education Personalization Educational Resource Recommendation Deep Learning
分类号:
G420
DOI:
10.15881/j.cnki.cn33-1304/g4.2021.05.005
文献标志码:
A
摘要:
知识追踪是通过跟踪历史学习轨迹,对学习者与学习资源间的学习交互过程进行建模,深入分析、挖掘、追踪学习者的动态知识掌握水平与认知结构,并准确预测未来的学习表现的一项智能技术,旨在实现人机协同优化教学过程、助力学习者美好发展的智慧教育愿景。为此,首先基于169篇国内外文献的分析,采用大数据建模、深度学习、多模态分析等方法,以智慧教育视域界定知识追踪为核心概念,从基于概率图的知识追踪、基于深度学习的知识追踪、基于参数估计的知识追踪这三个维度,探究其最新研究进展及面临的挑战;进而基于数据建模视角,从数据层、模型层、决策层、应用层这四个层面探讨知识追踪存在的问题,提出了相应的解决思路及应对策略,并进一步提出了智慧教育视域下的知识追踪框架,探讨了知识追踪在智慧教育中的应用路径及未来发展趋势。研究显示,智慧教育视域下的知识追踪能够更好地表征学习者复杂的学习状态,对其的研究可为实践智慧教育提供技术支撑。
Abstract:
Knowledge tracking is an intelligent technology that models the learning interaction process between learners and learning resources by tracking the historical learning track, deeply analyzes, mines and tracks learners’ dynamic knowledge mastery level and cognitive structure, and accurately predicts future learning performance. It aims to realize man-machine cooperation and optimize the teaching process The vision of wisdom education to help learners develop better. Therefore, firstly, based on the analysis of 169 domestic and foreign literatures, using the methods of big data modeling, deep learning and multimodal analysis, and taking the definition of knowledge tracking from the perspective of intelligent education as the core concept, this paper explores its latest research progress and challenges from the three dimensions of knowledge tracking based on probability map, knowledge tracking based on deep learning and knowledge tracking based on parameter estimation; Then, based on the perspective of data modeling, this paper discusses the existing problems of knowledge tracking from the four levels of data layer, model layer, decision-making layer and application layer, puts forward the corresponding solutions and countermeasures, further puts forward the knowledge tracking framework from the perspective of intelligent education, and discusses the application path and future development trend of knowledge tracking in intelligent education. Research shows that knowledge tracking from the perspective of wisdom education can better represent learners’ complex learning state, and its research can provide technical support for practical wisdom education.

备注/Memo

备注/Memo:
基金项目:本文系国家自然科学基金项目“细粒度的智能学习诊断及其可解释性机制研究”(项目编号:62177022)、国家自然科学基金项目“基于量化学习的在线教学视频评价模型与方法研究”(项目编号:62007011)、湖北省教育信息化研究中心开放基金项目“面向大型在线学习社区的学习监测及干预机制研究”(项目编号:HRCEI2020F0102)的研究成果。
更新日期/Last Update: 1900-01-01