国内数字化学习资源推荐算法研究热点、趋势和启示  

Research Hotspots,Trends and Implications of Recommendation Algorithms for Digital Learning Resources in China

作　　者：宋银李 孙瑜[1] SONG Yin-li;SUN Yu(School of Information,Yunnan Normal University,Kunming 650500,China)

机构地区：[1]云南师范大学信息学院,云南昆明650500

出　　处：《云南师范大学学报（自然科学版）》2022年第3期60-66,共7页Journal of Yunnan Normal University:Natural Sciences Edition

基　　金：国家自然科学基金资助项目(60903131);教育部科学计划研究重点资助项目(210210).

摘　　要：选取国内北大核心、CSSCI和CSCD核心期刊2007-2021年刊载的与数字化学习资源推荐算法相关的594篇文献,利用VOSviewer与CiteSpace进行可视化分析,发现国内数字化学习资源推荐算法的研究热点主要围绕技术应用、算法改进与优化、推荐策略设计、推荐效果评价及个性化推荐等方面,研究经历了萌芽初创、探索规划、发展应用、深入拓展及融合创新五个时期;提出我国未来数字化学习资源推荐算法发展需要从教育的视角来优化与改进;从算法推荐策略-推荐内容-学习真实情景等多维度结合,融合混合推荐、深度学习、社会化标注及情景感知等新兴技术,加强算法、模型和系统间动态联系等方面的思考.A total of 594 literatures related to the recommendation algorithm of digital learning resources published in the core journals of Peking University,CSSCI and CSCD from 2007 to 2021 were selected for visual analysis using VOSviewer and CiteSpace.It was found that the research hotspots of it in China mainly focused on technology application,algorithm improvement and optimization,recommendation strategy design,recommendation effect evaluation and personalized recommendation.The research went through five periods:budding,exploration and planning,development and application,in-depth expansion and fusion innovation.In view of this,the future development of digital learning resource recommendation algorithm in China needs to be optimized and improved from the perspective of education;from algorithm recommendation strategy-recommendation content-learning real situation and other multidimensional combination;integrating emerging technologies such as hybrid recommendation,deep learning,social tagging and situational awareness;strengthen the algorithm,model,dynamic relationship between systems and other aspects of thinking and inspiration.

关 键 词：数字化学习资源 推荐算法 可视化分析 VOSviewer CITESPACE 

分 类 号：G420[文化科学—课程与教学论]