大数据视角下的慕课评论语义分析模型及应用研究

发布时间：2018-04-10 15:42

  本文选题：慕课 + 课程评论　； 参考：《电化教育研究》2017年11期

【摘要】：文章从大数据的视角,应用语义分析的方法对慕课评论进行了分析和挖掘。针对慕课评论数量大且信息混杂的特点,文章提出了一种面向大数据的慕课评论语义分析模型。在该模型中,慕课评论被分为三种主要的类别:内容相关类、情感相关类和其他类。针对不同类别的差异,文章提出了基于词类的语义特征用于对评论进行表征和分类。以爱课程慕课上的四门课程评论作为实例进行分析发现:(1)以词类作为语义特征进行评论分类,单课程内部分类精度可达到84.36%,跨课程分类精度可达到79.72%以上;(2)针对内容相关类评论,通过词云分析可发现学习者的关注热点;(3)针对情感相关类评论,通过情感分析可评价学习者对课程的情感倾向;(4)针对其他类评论,通过关键词过滤和句式分析,可挖掘出学习者求助信息,完善课程支持服务。
[Abstract]:From big data's angle of view, this paper analyzes and excavates the comment of Mu class by semantic analysis.In view of the large number of reviews and mixed information, this paper puts forward a semantic analysis model for big data.In this model, MU reviews are divided into three main categories: content related, affective and other categories.According to the differences of different categories, this paper proposes a semantic feature based on part of speech for representation and classification of comments.Taking the four course reviews in Love course as an example, we find that the comments are classified by parts of speech as semantic features.The accuracy of single course classification can reach 84.36%, and the accuracy of cross-course classification can reach 79.72%. (2) through word cloud analysis, it can be found that learners pay more attention to the comments of affective related categories.Through affective analysis, we can evaluate learners' affective tendency to curriculum. (4) aiming at other kinds of comments, through keyword filtering and sentence structure analysis, we can find out learners' help information and perfect curriculum support service.
【作者单位】： 华中师范大学教育信息技术学院;浙江师范大学教师教育学院;中南民族大学教育学院;
【基金】：国家自然科学基金项目“网络学习资源深度聚合及个性化服务机制研究”(项目编号:71704062) 国家自然科学基金项目“非数学语言描述问题的机器理解方法研究”(项目编号:61772012)
【分类号】：G434
，

本文编号：1731837