首页 | 本学科首页   官方微博 | 高级检索  
     检索      

两种聚类算法在网站用户细分中的比较
引用本文:肖亚铁,柳亚飞,李鲁群,李哲敏.两种聚类算法在网站用户细分中的比较[J].上海师范大学学报(自然科学版),2018,47(1):49-52.
作者姓名:肖亚铁  柳亚飞  李鲁群  李哲敏
作者单位:上海师范大学信息与机电工程学院;上海师范大学音乐学院;
摘    要:给出了K-means算法和层次聚类算法在具体网站用户细分中准确率的比较,在细分网站用户这一类问题中,K-means算法在聚类准确率和处理速度上具有较大的优势,能够满足网站用户细分准确率的基本要求,其聚类准确率达到95%左右,且K-means算法处理速度比较快;层次聚类算法的处理速度较K-means算法慢,且其聚类准确率在处理大量用户数据时低于92%,这对于处理网站用户数据这类信息并不具备优势.

关 键 词:聚类算法  层次  用户细分  准确率
收稿时间:2016/6/21 0:00:00

Comparison of two clustering algorithms in website user segmentation
Xiao Yatie,Liu Yafei,Li Luqun and Li Zhemin.Comparison of two clustering algorithms in website user segmentation[J].Journal of Shanghai Normal University(Natural Sciences),2018,47(1):49-52.
Authors:Xiao Yatie  Liu Yafei  Li Luqun and Li Zhemin
Institution:The College of Information and Mechanical Engineering, Shanghai Normal University, Shanghai 200234, China,The College of Information and Mechanical Engineering, Shanghai Normal University, Shanghai 200234, China,The College of Information and Mechanical Engineering, Shanghai Normal University, Shanghai 200234, China and Music College, Shanghai Normal University, Shanghai 200234, China
Abstract:In this paper,we compare the accuracy of K-means algorithm and Hierarchical clustering algorithm in specific website user segmentation.Among the problems of subdividing website users,K-means algorithm has the advantage of accuracy and processing speed The advantages of this algorithm are that it can meet the basic requirements of website user segmentation accuracy,the clustering accuracy is about 95%,and K-means algorithm is faster;Hierarchical clustering algorithm is slower than K-means algorithm,And its clustering accuracy is less than 92% when processing a large amount of user data,which is not advantageous for processing information such as website user data.
Keywords:clustering algorithm  hierarchy  user segmentation  accuracy
本文献已被 CNKI 等数据库收录!
点击此处可从《上海师范大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《上海师范大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号