吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

改进DBSCAN聚类算法在电子商务网站评价中的应用

姜建华1,2,3, 杨玉免1,3, 边海燕1,3, 康嘉容1,3, 王丽敏1,3, 刘颖1,3   

  1. 1. 吉林财经大学 管理科学与信息工程学院, 长春 130117; 2. 吉林财经大学 物流产业经济与智能物流实验室, 长春 130117;3. 吉林财经大学  互联网金融省重点实验室, 长春 130117
  • 收稿日期:2014-12-16 出版日期:2016-03-26 发布日期:2016-03-23
  • 通讯作者: 姜建华 E-mail:jjh@jlufe.edu.cn

Application of ECommerce Sites Evaluation withImproved DBSCAN Clustering Algorithm

JIANG Jianhua1,2,3, YANG Yumian1,3, BIAN Haiyan1,3, KANG Jiarong1,3, WANG Limin1,3, LIU Ying1,3   

  1. 1. School of Management Science and Information Engineering, Jilin University of Finance and Economics, Changchun 130117, China; 2. Key Laboratory of Logistics Industry Economy and Intelligent Logistics, Jilin University of Finance and Economics, Changchun 130117, China; 3. Jilin Province Key Laboratory of Internet Finance, Jilin University of Finance and Economics, Changchun 130117, China
  • Received:2014-12-16 Online:2016-03-26 Published:2016-03-23
  • Contact: JIANG Jianhua E-mail:jjh@jlufe.edu.cn

摘要:

针对全国100家电子商务示范企业的相关数据, 先采用因子分析法对高维数据进行降维处理; 再通过改进DBSCAN(densitybased spatial clustering of applications with noise)算法对降维后的密度不均数据进行聚类分析, 得到了更合理的聚类结果; 最后根据聚类结果对相关示范企业提出改进建议.

关键词: 电子商务网站, 因子分析, DBSCAN算法, 聚类分析

Abstract:

In view of the relevant data of 100 ECommerce demonstration enterprises in China, we first used factor analysis method to reduce the dimension of high dimensional data, then we processed the uneven density data of reduced dimension by using improved DBSCAN (densitybased spatial clustering of applications with noise) algorithm, and obtained more reasonable clustering results. Finally, according to the clustering results, we put forward some suggestions to the relevant enterprises.

Key words: E-commerce sites, factor analysis, DBSCAN algorithm; cluster analysis

中图分类号: 

  • TP391