文献类型：期刊文献

中文题名：基于聚类算法的学生消费与成绩关联性分析

英文题名：Correlation analysis of students’consumption and score based on clustering algorithm

作者：蔡源[1];沈斌[1];房一泉[1]

机构：[1]华东理工大学,上海200237

年份：2023

卷号：47

期号：10

起止页码：106

中文期刊名：信息技术

外文期刊名：Information Technology

收录：CSTPCD

语种：中文

中文关键词：校园一卡通;关联性分析;数据挖掘;聚类算法;缺失值

外文关键词：campus card;correlation analysis;data mining;clustering algorithm;missing value

摘要：为了对学生一卡通数据、教务数据、图书馆数据进行相关性研究,挖掘这些数据的隐藏价值,文中提出基于聚类算法的学生消费数据与成绩数据的关联性分析。首先对原始数据进行预处理,使用Cubic插值法对缺失值进行插补处理,并采用主成分分析法将原始数据降至一维。然后通过改进的聚类算法K-Means++算法对降维后的数据进行了初步的聚类,并对不同k值取得的聚类结果进行性能比较,最后对8287个实验样本进行关联性计算。所得出实验的结果可以为学校提供决策帮助。
In order to study the correlation of student campus card data,educational data and library data,and mine the hidden value of these data,this paper proposes a correlation analysis between students’consumption data and score data based on clustering algorithm.Firstly,the original data are preprocessed and the missing values are interpolated by cubic interpolation algorithm,then the original data are reduced to one dimension by principal component analysis method.Then,the reduced dimension data are preliminarily clustered by the improved clustering algorithm K-Means++,and the clustering results obtained by different kvalues are compared.Finally,the correlation analysis of 8287 experimental samples is calculated.The experiment results can provide decision-making help for schools.