基于K-medoids聚类的层次化教学质量提升评估研究

Research on hierarchical teaching quality improvement evaluation based on K-medoids clustering

研究基于K-medoids聚类的层次化教学质量提升评估方法。通过K-medoids聚类分析数据样本间的最优路径和聚类中心,获取数据样本集的历史最优位置,将历史最优位置点看成不同评估指标;以教学队伍、教学内容、教学条件等两级教学质量评估指标体系作为评估指标构建教学质量评估指标体系;采用多层次评价模型对历史最优位置点,即评估指标实行层次分析,通过分层综合评估过程先对评估指标实施一级评估,确定各评估指标权重,依据权重构建教学质量效果判断矩阵;再采用判断矩阵完成二级评估,评估层次化教学质量。实验结果表明,该方法评估计算机学院教师教学质量为较高,且二班教师教学质量优于一班教学质量。

The hierarchical teaching quality improvement evaluation method based on K-medoids clustering is studied. The optimal path and clustering center among the data samples are analyzed by K-medoids clustering to obtain the historical optimal position of the data sample set,and the historical optimal position points are regarded as different evaluation indexes. The teaching quality evaluation index system is constructed by taking the two-level teaching quality evaluation index system(consisting of teaching team,teaching content and teaching conditions) as evaluation indexes. Hierarchical analysis on the historical optimal position points,that is,the evaluation indexes is implemented by multi-level evaluation model. Firstly,the evaluation indexes are subject to the first-order evaluation to determine the weights of each evaluation index by hierarchical integrated evaluation process,and the teaching quality effect judgment matrix is constructed based on the weights. Then the judgment matrix is used to complete the second-order evaluation,named the hierarchical teaching quality evaluation. The experimental results show that the teaching quality of the teachers in Computer College is higher,and the teaching quality of the teachers in Class 2 is better than that of the teachers in Class 1.