摘要
自适应学习是个性化学习的实现方式,通过分析收集到的学习者实时交互数据来自主引导学习者学习。目前自适应学习还处在弱人工智能水平,而实现自适应学习向强人工智能水平转变的突破口在于自适应学习算法的突破。同时,自适应学习算法是实现自适应学习的关键支撑,因此,有必要对自适应学习算法及其应用进展进行深入分析。本研究首先梳理了当前主流自适应学习平台使用的自适应学习算法,将自适应学习算法归为心理测量类和机器学习类,并详细介绍了这几类算法的具体应用,从而总结出各类算法的优缺点,最后提出将两类算法进行结合,根据优势互补原则组合多种自适应学习算法的应用策略和建议。
Adaptive learning,as a key support for personalized learning,is used to guide student’s self-learning by collecting and analyzing data from learners real-timely.At present,adaptive learning remains at the level of weak artificial intelligence,while the breakthrough towards the level of strong artificial intelligence depends on the advances of adaptive learning algorithms.Meanwhile,the implementation of adaptive learning relies heavily on adaptive learning algorithms.Therefore,it is necessary to conduct an in-depth analyze of adaptive learning algorithms and their application progress.This study firstly combs the adaptive learning algorithms used by the current mainstream adaptive learning platform,classifies the adaptive learning algorithms into psychometrics category and machine learning category,and introduces the specific application of these kinds of algorithms in detail,thus conclude the advantages and disadvantages of various algorithms.This study finally provides suggestions on deep integrations of adaptive learning algorithms and suggestions on combines multiple adaptive learning algorithms according to the principle of complementary advantages.
作者
汪存友
赵燕飞
王亚青
WANG Cunyou;ZHAO Yanfei;WANG Yaqing(School of Communication,Shanxi Normal University,Linfen 041004,China;School of Educational Science,Shanxi Normal University,Linfen 041004,China)
出处
《开放学习研究》
2020年第2期40-46,共7页
Journal of Open Learning
基金
2019年度教育部人文社科研究青年基金项目“认知诊断理论在个性化学习资源推荐系统中的应用研究”(项目编号:19YJC880080)的阶段性成果。
关键词
自适应学习
算法
机器学习
心理测量学
自适应学习平台
adaptive learning
algorithms
machine learning
psychological measurement
adaptive learning platforms
