计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (17): 169-179.DOI: 10.3778/j.issn.1002-8331.1902-0223

• 模式识别与人工智能 • 上一篇    下一篇

基于聚类和关联算法的学生成绩挖掘与分析

郭鹏,蔡骋   

  1. 西北农林科技大学 信息工程学院,陕西 杨陵 712100
  • 出版日期:2019-09-01 发布日期:2019-08-30

Data Mining and Analysis of Students’ Score Based on Clustering and Association Algorithm

GUO Peng, CAI Cheng   

  1. College of Information Engineering, Northwest A&F University, Yangling, Shaanxi 712100, China
  • Online:2019-09-01 Published:2019-08-30

摘要: 针对某高校信息工程学院学生的学习状况和培养方案的改进需求,以2008—2014级信息工程学院本科生课程成绩为研究对象,提出一种基于改进K-means和引入兴趣度的Apriori的学生课程成绩分析方法。采用改进的K-means算法对成绩信息进行离散化处理,采用引入兴趣度的Apriori算法进行挖掘并根据得到的课程之间的关联规则绘制课程关系网络图,对课程间的关联关系、衔接关系以及课程的重要程度进行分析。应用所述方法进行挖掘能够减少大量没有意义的规则,提高了挖掘结果的准确性,挖掘所得到的结果不仅能够为教学方案的设计和改进提供一定的参考信息,还有助于提高学校的教学质量和学生的学习质量。

关键词: 数据挖掘, 成绩分析, 关联规则, 聚类

Abstract: In order to meet the needs of the improvement of the students’ learning status and training scheme reform in the college of information engineering of a university, taking the course scores of undergraduates in the college of information engineering from 2008 to 2014 as the objects of study, a method which combines the Apriori algorithm with interest and the improved K-means is proposed to analyze students’ grades. First of all, this research uses the improved K-means to discretize the scores. Then, Apriori algorithm with the degree of interest is adopted for data mining. According to the association rules between courses and the drawing of network diagram of course relations, this research analyzes the relationship among courses and the importance of courses. Finally, the method can reduce a lot of meaningless rules and improves the accuracy of mining results. And the results obtained by using the method can not only provide some reference information for the design and improvement of teaching programs, but also help to improve the teaching quality of schools and students’ learning quality.

Key words: data mining, score analysis, association rule, clustering