【作 者】：
吴林静， 高 喻， 涂凤娇， 王瑾洁， 刘清堂
【关 键 词 】：
在线协作会话； 学习投入； 语义分析； 聚类分析
【栏 目】：
课程与教学
【中文摘要】：
在线协作会话是在线学习中用于促进交流、实现协同知识建构的重要手段。对在线协作会话的自动分析能够帮助教师了解学习者的学习状态，提升在线教育质量。文章针对在线协作会话的特征，提出了五维度的在线协作会话分析框架：认知投入、情感投入、行为投入、社交投入和感知投入，并给出了各维度的度量指标和计算方法。在该框架的基础上，文章以话语的语义心理特征为基础，进一步提出了基于语义的协作会话学习投入自动分析模型，以实现学习投入的自动化分析。以“现代教育技术”课程中的在线协作会话过程为例进行了案例分析。案例分析发现学习者的学习投入存在五种典型模式：“三好学生型”“勤奋型”“认知型”“中规中矩型”和“低成就型”学习者。基于不同的学习投入模式，为教师给出了不同的教学策略建议。案例研究验证了自动分析模型的有效性和可行性，为在线协作会话过程的自动量化分析提供了新的方法和思路。
【英文摘要】：
The online collaborative conversation is an important means to promote communication and realize collaborative knowledge building in online learning. Automatic analysis of online collaborative conversations can help teachers understand the learning status of learners and improve the quality of online education. This paper proposes a five-dimensional analysis framework for online collaborative conversations based on the characteristics of online collaborative conversations: cognitive engagement, emotional engagement, behavioral engagement, social engagement and perceptual engagement, and gives the measurement indexes and calculation methods for each dimension. Based on this framework and the semantic psychological features of discourse, this paper further proposes an automatic semantic-based analysis model to realize the automatic analysis of learning engagement. An online collaborative conversation in the course of "Modern Educational Technology" is taken as an example. It is found that there are five typical patterns of learners' learning engagement: high engagement high output, high engagement medium output, high cognitive engagement medium output, medium engagement medium output, and low engagement low output. Based on different learning engagement patterns, different teaching strategies are suggested for teachers. The case study verifies the validity and feasibility of the automatic analysis model and provides a new method and idea for the automatic quantitative analysis of online collaborative conversations.