摘要
个性化学习路径推荐是智能技术驱动教育服务智能升级的关键力量,是实现大规模个性化教育的重要驱动。然而,目前个性化学习路径推荐的研究与实践仍未成熟,难以满足学习者因人而异、因时而变的个性化需求。个性化学习路径推荐的主流推荐框架包括基于机器学习、基于进化计算和基于知识图谱三种方式,具有不同的模型机理和适用的教育场景,经历了起步探索、预测推理、改进优化和融合创新的技术演进历程。其对教育的价值导向体现在赋能课堂教学、助力因材施教,优化在线学习、驱动服务升级,支持场馆学习、增强智能感知,改善游戏学习、优化交互体验,为破解大规模与个性化相结合的应用难题提供了重要支撑。随着智能技术的迭代升级及深入应用,未来个性化学习路径推荐呈现四大发展趋势,即关注信息要素的深度融合、迈向全方位多尺度的形式化建模,融合多种技术优势、构建以知识为主导的高性能推荐框架,注重可视化呈现和动态感知、推动个性化服务模式的优化升级,重视全景性应用策略研究、促进推荐技术与真实教育情境的紧密耦合。
Personalized learning path recommendation is a key driving force for intelligent upgrading of education services driven by intelligent technology,and it is an important driver for achieving large-scale personalized education.However,the research and practice of personalized learning path recommendation are still immature,making it difficult to meet the personalized needs of learners that vary from person to person and over time.The mainstream recommendation frameworks for personalized learning paths include three methods:machine learning-based,evolutionary computation-based,and knowledge graph-based.These have different model mechanisms and applicable educational scenarios,and have undergone a technological evolution process of initial exploration,prediction and reasoning,improvement and optimization,and fusion and innovation.Its value orientation towards education is reflected in empowering classroom teaching,helping tailor instruction,optimizing online learning,driving service upgrades,supporting venue learning,enhancing intelligence perception,improving game-based learning,optimizing interactive experiences,and providing important support for solving the difficult problem of combining large-scale and personalized applications.With the iterative upgrading and in-depth application of intelligent technology,future personalized learning path recommendation will present four major development trends,namely,focusing on the deep integration of information elements,moving towards all-round multi-scale formal modeling,integrating multiple technical advantages,constructing a high-performance recommendation framework dominated by knowledge,emphasizing visual presentation and dynamic perception,promoting the optimization and upgrading of personalized service modes,attaching importance to panoramic application strategy research,and promoting the close coupling of recommendation technology and real education situations.
作者
郑雅倩
李新
李艳燕
王德亮
包昊罡
ZHENG Yaqian;LI Xin;LI Yanyan;WANG Deliang;BAO Haogang(Beijing Normal University,Beijing 100089;Jiangsu Normal University,Xuzhou Jiangsu 221116;The University of Hong Kong,Hong Kong 999077,china;The China National Academy of Educational Sciences,Beijing 100088)
出处
《现代远距离教育》
CSSCI
2023年第3期39-47,共9页
Modern Distance Education
基金
国家自然科学基金面上项目“融合多模态学习分析的协作过程监测和智能反馈研究”(编号:62277006)
北京市自然科学基金面上项目“面向数字社会发展的智慧教育支持服务关键技术研究”(编号:9222019)。
关键词
个性化学习路径推荐
模型机理
技术演进
价值导向
发展趋势
Personalized Learning Path Recommendation
Model Mechanism
Technology Evolution
Value Orientation
Development Trend
作者简介
郑雅倩,北京师范大学教育技术学院博士研究生;李新,博士,江苏师范大学智慧教育学院讲师;通信作者:李艳燕,博士,北京师范大学教育技术学院教授,博士生导师;王德亮,中国香港大学教育学院博士研究生;包昊罡,博士,中国教育科学研究院基础教育研究所助理研究员。
