摘要
以学习者学习行为为研究对象,提取了学习行为特征,构建了一组学习行为时间序列数据,提出了一种学习行为时间序列相似性模型;通过学习行为相似性模式的表示、度量和聚类,验证模型的有效性;结果表明,能够很好地对学习行为进行分类,对无效样本有效检验。
The object of this paper is that learners'learning behavior is,and the learning behavior characteristics are extracted,built a set of learning behavior time series data,and puts forward to a similarity model based on learning behavior time series.The validity of the model is verified by the representation,measurement and clustering of learning behavior similarity patterns.The experimental results show that the learning behavior can be classified well and the invalid specimen can be effectively classified.
作者
姜廷慈
李敬有
吕洪柱
JIANG TING-ci;LI JING-you;LV HONG -zhu(School of Computer and Control,Qiqihar University,Heilongjiang Qiqihar 161006,China)
出处
《齐齐哈尔大学学报（自然科学版）》
2019年第6期1-3,7,共4页
Journal of Qiqihar University(Natural Science Edition)
基金
黑龙江省教育厅科学技术研究项目(12541872)
关键词
多元时间序列
学习行为
相似性度量
动态时间弯曲距离
multivariate time series
learning behavior
similarity measure
dynamic time warping distance
