计算机科学 ›› 2017, Vol. 44 ›› Issue (9): 58-61.doi: 10.11896/j.issn.1002-137X.2017.09.011

• CRSSC-CWI-CGrC 2016 • 上一篇    下一篇

一种基于非负矩阵分解的聚类集成算法

何梦娇,杨燕,王淑营   

  1. 西南交通大学信息科学与技术学院 成都610031,西南交通大学信息科学与技术学院 成都610031,西南交通大学信息科学与技术学院 成都610031
  • 出版日期:2018-11-13 发布日期:2018-11-13
  • 基金资助:
    本文受国家自然科学基金项目(61572407),国家科技支撑计划课题项目(2015BAH19F02)资助

NMF-Based Clustering Ensemble Algorithm

HE Meng-jiao, YANG Yan and WANG Shu-ying   

  • Online:2018-11-13 Published:2018-11-13

摘要: 为了解决通过原始数据集获得的基聚类结果存在一定的信息丢失,从而使得集成阶段的有效信息减少的问题,提出了一种基于非负矩阵分解的K-means聚类集成算法。该算法先利用K-means聚类算法获得集成信息矩阵,然后从原始数据集获取数据相关性,将两者结合后通过非负矩阵分解(NMF)技术构建共识函数以获得最终结果。实验证明,所提算法可以有效获取原始数据的潜在信息,并提高聚类质量。

关键词: 聚类集成,K-means,NMF,潜在信息

Abstract: A NMF-based K-means clustering ensemble (NBKCE) algorithm was proposed for solving the problem of effective information loss in ensemble,which is caused by basic clustering results obtained from the original datasets.In NBKCE,an ensemble information matrix is built primarily by exploiting the results of the K-means,and then the relationship matrix is formed based on the original dataset.At last nonnegative matrix factorization (NMF) is employed to construct consensus function to gain the final results.The experiments demonstrate that the NBKCE may attain the underlying information effectively and improve the performance of the clustering.

Key words: Ensemble clustering,K-means,NMF,Underlying information

[1] YANG C Y,LIU D Y,YANG B,et al.The research on clustering ensemble[J].Computer Science,2011,8(2):166-170.(in Chinees) 杨草原,刘大有,杨博,等.聚类集成方法研究[J].计算机科学,2011,38(2):166-170.
[2] STREHL A,GHOSH J.Cluster ensembles-a knowledge reuse framework for combining multiple partitions[J].Journal of Machine Learning Research,2003,3(3):583-617.
[3] WANG H J,LI Z S,CHENG Y,et al.A Latent Variable Model for Cluster Ensemble[J].Journal of Software,2009,20(4):825-833.(in Chinees) 王红军,李志蜀,成飏,等.基于隐含变量的聚类集成模型[J].软件学报,2009,20(4):825-833.
[4] ZHOU Z H.Ensemble Methods:Foundations and Algorithms[M].Taylor & Francis,2012.
[5] YANG Y,KAMEL M.An aggregated clustering approach using multi-ant colonies algorithms[J].Pattern Recognition,2006,38(7):1278-1289.
[6] LAMON N,BOONGOEN T,GARRETT S.Link-based cluster ensemble approach for categorical data clustering[J].IEEE Transactions on Knowledge and Data Engineering,2012,4(3):413-425.
[7] YANG Y,FENG C F,JIA Z,et al.A Link-Based Fuzzy Clustering Ensemble[J].Journal of University of Electronic Science and Technology of China,2014,3(6):887-892.(in Chinees) 杨燕,冯晨菲,贾真,等.基于链接的模糊聚类集成方法[J].电子科技大学学报,2014,3(6):887-892.
[8] HAN J,KAMBER M.Data Mining:Concepts and Techniques[J].Data Mining Concepts Models Methods & Algorithms Se-cond Edition,2006,5(4):1-18.
[9] DING C,HE X,SIMON H.Nonnegative lagrangian relaxationof k-means and spectral clustering[C]∥ECML.2005: 530-538.
[10] ZHANG J S,WANG C P,YANG Y Q.Learning latent features by nonnegative matrix factorization combining similarity judgments[J].Neurocomputing,2015,155:43-52.
[11] MIAO L D,QI H R.Endmember extraction from highly mixed data using minimum volume constrained nonnegative matrix factorization[J].IEEE Trans.Geosci.Remote Sens.,2007,45(3):765-777.
[12] ASUNCION A,NEWMAN D J.UCI machine learning repository school of information and computer science,university of california [DB/OL].(2007-06-02).http://www.ics.uci.edu/-m-learn/MLRepository.html.
[13] FERN X Z,BRODLEY C E.Solving cluster ensemble problems by bipartite graph partitioning[C]∥Proc.21th Int.Conf.Mach.Learn..2004:36-44.
[14] ALEXANDER T,ANIL K J,WILLIAM P.Clustering Ensem-bles:Models of Consensus and Weak Partitions[J].IEEETrans.on Pattern Analysis and Machine Intelligence,2005,27(12):1866-1881.
[15] YANG Y,JIN F,KAMEL M.Survey of clustering validity eva-luation[J].Application Research of Computer,2008,25(6):1630-1632.(in Chinese).杨燕,靳蕃,KAMEL M.聚类有效性评价综述[J].计算机应用研究,2008,25(6):1630-1632.
[16] RAND W M.Objective criteria for the evaluation of clustering methods[J].Journal of American Statistical Association, 1971,66(336):846-850.

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!