计算机与现代化

• 算法设计与分析 • 上一篇    下一篇

基于均值聚类和决策树算法的学生成绩分析

  

  1. 广东科学技术职业学院,广东珠海519090
  • 收稿日期:2014-03-06 出版日期:2014-06-13 发布日期:2014-06-25
  • 作者简介:龙钧宇(1979-),男,湖南双峰人,广东科学技术职业学院讲师,硕士,研究方向:信号与信息处理,模式识别。
  • 基金资助:
     广东科学技术职业学院校级科研项目(XJZD201201)

Analysis of Student’s Achievements Based on Mean Cluster and Decision Tree Algorithm

  1. Guangdong Vocational Institute of Technology, Zhuhai 519090, China
  • Received:2014-03-06 Online:2014-06-13 Published:2014-06-25

摘要:  针对高职学生的学习情况,采用k-均值聚类算法对学生的考试成绩进行等级划分,再采用R-C4.5算法构造决策树,通过对该决策树提取规则来分析学生各学科成绩和总评成绩的相关性。该方法可以减少决策树中的无意义的分支,挖掘出影响学生总评成绩的主要因素,为任课教师和教学管理人员在制定教学计划、开展教学工作和进行教学评价等方面提供参考。

关键词:  , 成绩相关性, k-均值聚类, 决策树, R-C4.5算法

Abstract: According to the achievements of the vocational students, k-means cluster analysis algorithm is applied to divide the degrees of the student’s achievements, then the R-C4.5 algorithm is applied to build the decision tree, finally rules are extracted by the decision tree to analyze the relationship between the achievement of each subject and the student’s final total achievement. By this algorithm the meaningless branch of the decision tree can be reduced, and the main fator which affecting the student’s total achievement can be mined effectively, thus providing references to teachers and teaching managers to make teaching plans, carrying out teaching activities, and evaluating the teaching results.

Key words: achievement relativity, k-means cluster analysis algorithm, decision tree, R-C4.5 algorithm