基于模糊聚类的CSCL学习者混合属性分组

摘要

计算机支持的协作学习研究大部分只依据个别定量属性计算学习者之间的差异度,进而根据某种算法对学习者进行分组,分组结果并不能反映学习者本质差异.提出一种综合学习者定性和定量属性的混合属性来计算差异度,使用模糊聚类算法对学习者进行分组的方法.实验表明该方法简单易行,分组结果较为准确.%Most of researches on computer supported collaborative learning calculate the dissimilarity between learners only relying on individual quantitative attribute, and then group the learners according to a certain algorithm, but the grouping result can not reflect the essential dissimilarity of learners. A new approach is proposed in the paper. It calculates the dissimilarity with mixed attributes that combines the qualitative and quantitative attributes of the learners, and groups the learners with fuzzy clustering. Experiment shows, the new approach is simple to implement and has preferable accuracy in grouping result.