针对网络学习空间中的数据未能被充分挖掘和利用的问题,文章提出了一种基于数据挖掘技术的在线学习行为分析模型.该模型将网络学习空间中的在线学习行为分为四类:独立学习行为、系统交互行为、资源交互行为和社会交互行为.基于此四类行为,该模型提出了相关分析、分类分析和聚类分析的分析方法,依据分析结果为网络学习中的各类利益相关者提供教学参考.文章以某网络学习空间数据为例,进行案例研究后发现:①独立学习行为与学习成绩的相关性最强,具有较强的预测作用;学习者的求助次数与其它因素相关性最弱,不宜作为行为分析的核心指标;②以具有较强相关性的学习行为对成绩进行预测,使用 K-近邻分类器可获得84.62％的准确率;③对学习行为进行聚类分析,可发现学习者主要存在四种不同的学习行为模式,针对不同类型的学习者,教师可采取不同的干预策略,实现个性化的教与学.%For the problem that the data in network learning space is not fully tapped and utilized,this article proposed an online learning behavior analysis model based on data mining technology.In this model,online learning behaviors were classified into four categories: independent learning behavior,system interaction behavior,resource interaction behavior and social interaction behavior.Based on these four categories,correlation analysis,classification analysis and clustering analysis were applied to analyze the data and provide suggestions for the stakeholders in network learning space.Taking data from one network learning space as a case,we found that: ①Independent learning behaviors were strong predictors and had the strongest correlation with learning performance.The frequency of asking for help had the weakest correlation with other behaviors,and it was not suitable to consider it as a key indicator in analyzing learner behaviors.②The accuracy of performance prediction was 84.62％with K-nearest Neighbor Classification based on the strong correlated learning performance.③clustering analysis indicated that learners could be divided into four typical categories in terms of learning behavior model.As a result,different strategies could be offered to students of different types to promote individualized teaching and learning.
展开▼
