数据驱动的多维融合在线协作会话可计算模型及应用研究

Research on the Computable Model and Application of Data-driven Multi-dimensional Fussed Online Collaborative Discourse

在线协作会话是在线学习活动的重要组成部分,也是促进学习者进行协同知识建构的重要手段,因此越来越受关注。目前,由于分析框架和分析方法不够完善,导致当前在线协作会话存在分析维度单一、自动化程度较低等问题。为此,文章首先提出了一种数据驱动的多维融合在线协作会话分析框架和可计算模型,然后提出了协作水平系数和协作投入倾向两个核心指标对会话状态进行整体评估,最后进行了案例分析。案例分析的结果表明:协作水平系数能够准确表征学习者在在线协作会话中的协作状态;在线协作会话可计算模型可以直观地呈现学习者在线协作会话过程中的核心特征和学习者的个体差异性;学习者的在线协作会话模式包括领导型、思想型、中庸型和边缘型四种典型类型。文章通过研究验证了模型的有效性和其对于在线协作会话自动分析与评估的实践价值,期望能够为在线协作会话的自动化分析提供参考。

As an important part of online learning activities,online collaborative discourse is also an important method to promote learners to carry out collaborative knowledge construction,which has attracted more and more attention.However,due to not enough perfect analysis framework and analysis methods,the current online collaborative discourse exists the problems such as single analysis dimension and low degree of automation.Therefore,this paper firstly proposed a computable model of data-driven and multi-dimensional fused online collaborative discourse,and then put forward to evaluate the overall state of of collaborative discourse by tow core indicators of collaboration level coefficient and collaboration engagement tendency,and finally made a case study.The results of the case study showed that collaboration level coefficient could accurately represent learners’collaborative state in collaborative discourse,and the computable model of online collaborative discourse could intuitively present learners’core characteristics and individual differences in the process of online collaborative discourse.Meanwhile,learners’online collaborative discourse model included four types of leadership type,thought type,moderation type and marginal type.Through research,the effectiveness of the model and its practical value for automatic analysis and evaluation of online collaborative discourse were verified in this paper,which was expected to provide reference for the automatic analysis of online collaborative discourse.