首页 | 本学科首页   官方微博 | 高级检索  
     

一种新的基于局部保持投影的高维数据聚类成员构造方法
引用本文:周静波,殷俊,金忠.一种新的基于局部保持投影的高维数据聚类成员构造方法[J].计算机科学,2011,38(9):177-181.
作者姓名:周静波  殷俊  金忠
作者单位:(南京理工大学计算机科学与技术学院 南京 210094)
基金项目:本文受国家自然科学基金(60632050,60873151)资助。
摘    要:研究在高维数据中如何产生聚类成员,并提出一种新的构造聚类成员的方法。为解决高维数据的维度对构造成员带来的影响,新的构造方法在构造聚类成员之前利用局部保持投影先对高维数据进行维度约减,然后在约减后的子空间中用随机投影结合K均值方法构造聚类成员。最后讨论了局部保持投影子空间维度的选取。实验表明,新方法得到的结果要明显优于已有的主分量分析结合下采样方法和简单的随机投影方法。

关 键 词:聚类融合,维度约减,局部保持投影,随机投影

New Ensemble Constructor Based on Locality Preserving Projection for High Dimensional Clustering
ZHOU Jing-bo,YIN Jun,JIN Zhong.New Ensemble Constructor Based on Locality Preserving Projection for High Dimensional Clustering[J].Computer Science,2011,38(9):177-181.
Authors:ZHOU Jing-bo  YIN Jun  JIN Zhong
Affiliation:(School of (}c>mputer Science and Technology,Nanjing University of Science and Technology,Nanjing 210094,China)
Abstract:This paper studied how to construct cluster ensembles for high dimensional data and proposed a new ensemble constructor. ho ameliorate the effect caused by high dimensionality, the proposed method used Locality Preserving Projections(LPP) to reduce the dimensionahty before constructing ensembles. Then constructed ensembles based on random projection combined with K means in LPP subspace. Finally,we discussed how to choose the dimensionality of LPP subspace. hhe experiments show that ensembles generated by new algorithms perform better than those by Princi- pal Component Analysis with subsampling(PCASS) and simple Random Projection(RP) that was proposed before.
Keywords:Cluster cnsembles  Dimension reduction  I_ocality preserving projections  Random projection
本文献已被 CNKI 等数据库收录!
点击此处可从《计算机科学》浏览原始摘要信息
点击此处可从《计算机科学》下载全文
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