【作 者】：
张 治， 杨 熙， 夏冬杰
【关 键 词 】：
学习者画像； 学习行为投入； 在线作业； 测量指标； 学习分析
【栏 目】：
课程与教学
【中文摘要】：
学习者画像是描述学习者特征、实现智能化推送、实施个性化教育的重要基础。如何挖掘并利用在线学习平台中的数据构建学习者画像是当前亟待解决的问题。研究以在线作业为目标场景，以在线学习行为投入为切入点，构建了以参与、坚持、专注、学术挑战和自我调控为主要维度的分析框架和测量指标，利用7695名小学生在线作业数据进行了27个测量指标的有效性验证，采用K-Means聚类方法对在线学习者的行为特征和结果特征进行了标签分类，形成了四类学习者的群体画像，提出了相应的学习指导建议。研究发现，学习者的学业成绩与作业行为投入之间存在显著相关，不同的行为投入平台指标与学业成绩呈现不同的相关性，学习品质相关指标与学业成绩呈现强相关。因此，在线教育平台应通过画像技术，持续跟踪学习者的在线学习行为投入，评估学习者的学习品质，提出个性化的指导建议，推送精准化的学习资源，进而提升学习者在线学习效率。
【英文摘要】：
Learner profile is an important basis for describing the characteristics of learners, achieving intelligent push and implementing personalized education. How to mine and use the data of online learning platform to build the learner profile is an urgent problem to be solved. Taking online homework as the target scenario and online learning behavioral engagement as the entry point, this paper constructs an analysis framework and measurement indicators with participation, persistence, concentration, academic challenge and self-regulation as the main dimensions. The validity of 27 measurement indicators is verified by using the online homework data of 7695 primary school students. A K-Means clustering approach is used to classify the behavioral characteristics and outcome characteristics of online learners. A group profiles of four types of learners are formed, and corresponding learning recommendations are put forward. It is found that there is a significant correlation between learners' academic performance and homework behavioral engagement, different behavioral engagement platform indicators show different correlations with academic performance, and learning quality indicators are strongly correlated with academic performance. Therefore, online education platforms should continuously track learners' online learning behavioral engagement, assess learners' learning quality, provide personalized guidance and suggestions, and push precise learning resources through the profile technology, so as to improve learners' online learning efficiency.