基于学习者个性行为分析的学习结果预测框架设计研究

The Design Research of Learning Outcomes Prediction Based on the Model of Personalized Behavior Analysis for Learners

对学习者及其所在情境中产生的数据进行记录和分析并为其提供个性化学习服务是教育大数据背景下学习分析研究在促进有效学习方面的内在旨趣，而要实现对学习状态的评估、预警和干预则需要对学习者个性化信息进行实时建模。该文通过对国内外学习分析模型的梳理分析总结存在问题，并基于个性化学习模式和相关文献构建了数字化学习环境下个性化行为分析模型。在此基础上，以教学目标分类理论、个性化学习理论和社会认知理论为指导设计了“目标—过程—结果”为思想的学习结果分类设计原理，并对各层级的学习行为类型进行分析。最后基于学习行为分析模型和学习结果分类设计了学习结果预测框架，包括学习内容分析、学习行为分析和学习预测分析三个模块，为后面个性化学习分析工具的设计提供理论指导。

It is internal purport for recording and analysis of the data that is generated by learners and their contexts and providing personalized learning service in the aspect of promoting effective learning under the educational background of big data. The real time modeling is required to achieve the assessment of learning status， early warning as well as intervention. The existing problems were summarized by combing analysis for learning analysis model at domestic and abroad in this paper. And we built a personalized behavior model applying in the digital environment based on the personalized learning patterns and related literature. In addition， we designed learning outcome classification principles with the idea of “target-process-results” in the guidance of instruction theories which included teaching target classification theory， personalized learning theory and social cognitive theory. The type of behaviors was also analyzed at all levels. Finally， the learning outcomes prediction framework was designed based on learning behavior analysis model and learning results. The framework included learning content analysis， learning behavior analysis and learning prediction analysis， which could provide theoretical guidance for the design of personalized learning analysis tools.