A Unified Framework for Clustering Constrained Data Without Locality Property

被引:7
|
作者
Ding, Hu [1 ]
Xu, Jinhui [2 ]
机构
[1] Univ Sci & Technol China, Sch Comp Sci & Technol, Hefei, Peoples R China
[2] SUNY Buffalo, Dept Comp Sci & Engn, Buffalo, NY USA
关键词
Constrained clustering; k-means/median; Approximation algorithms; High dimensions;
D O I
10.1007/s00453-019-00616-2
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
In this paper, we consider a class of constrained clustering problems of points in R-d, where d could be rather high. A common feature of these problems is that their optimal clusterings no longer have the locality property (due to the additional constraints), which is a key property required by many algorithms for their unconstrained counterparts. To overcome the difficulty caused by the loss of locality, we present in this paper a unified framework, called Peeling-and-Enclosing, to iteratively solve two variants of the constrained clustering problems, constrained k-means clustering (k-CMeans) and constrained k-median clustering (k-CMedian). Our framework generalizes Kumar et al.'s (J ACM 57(2):5, 2010) elegant k-means clustering approach from unconstrained data to constrained data, and is based on two standalone geometric techniques, called Simplex Lemma andWeaker Simplex Lemma, for k-CMeans and k-CMedian, respectively. The simplex lemma (or weaker simplex lemma) enables us to efficiently approximate the mean (or median) point of an unknown set of points by searching a small-size grid, independent of the dimensionality of the space, in a simplex (or the surrounding region of a simplex), and thus can be used to handle high dimensional data. If k and 1/epsilon are fixed numbers, our framework generates, in nearly linear time (i.e., O(n(log n)(k+1)d)), O((log n)(k)) k-tuple candidates for the k mean or median points, and one of them induces a (1 + epsilon)-approximation for k-CMeans or k-CMedian, where n is the number of points. Combining this unified framework with a problem-specific selection algorithm (which determines the best k-tuple candidate), we obtain a (1 + epsilon)-approximation for each of the constrained clustering problems. Our framework improves considerably the best known results for these problems. We expect that our technique will be applicable to other variants of k-means and k-median clustering problems without locality.
引用
收藏
页码:808 / 852
页数:45
相关论文
共 50 条
  • [1] A Unified Framework for Clustering Constrained Data Without Locality Property
    Hu Ding
    Jinhui Xu
    Algorithmica, 2020, 82 : 808 - 852
  • [2] Locality Constrained Transitive Distance Clustering on Speech Data
    Liu, Wenbo
    Yu, Zhiding
    Raj, Bhiksha
    Li, Ming
    16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5, 2015, : 2917 - 2921
  • [3] A Unified Framework for Approximating and Clustering Data
    Feldman, Dan
    Langberg, Michael
    STOC 11: PROCEEDINGS OF THE 43RD ACM SYMPOSIUM ON THEORY OF COMPUTING, 2011, : 569 - 578
  • [4] A unified framework for approximating and clustering data
    California Institute of Technology, Pasadena, CA 91125, United States
    不详
    Proc. Annu. ACM Symp. Theory Comput., (569-578):
  • [5] A Unified Framework for Privacy Preserving Data Clustering
    Li, Wenye
    NEURAL INFORMATION PROCESSING (ICONIP 2014), PT I, 2014, 8834 : 319 - 326
  • [6] A unified framework for privacy preserving data clustering
    Li, Wenye
    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2014, 8834 : 319 - 326
  • [7] Weighted sparse simplex representation: a unified framework for subspace clustering, constrained clustering, and active learning
    Hankui Peng
    Nicos G. Pavlidis
    Data Mining and Knowledge Discovery, 2022, 36 : 958 - 986
  • [8] Weighted sparse simplex representation: a unified framework for subspace clustering, constrained clustering, and active learning
    Peng, Hankui
    Pavlidis, Nicos G.
    DATA MINING AND KNOWLEDGE DISCOVERY, 2022, 36 (03) : 958 - 986
  • [9] A Semi-NMF-PCA Unified Framework for Data Clustering
    Allab, Kais
    Labiod, Lazhar
    Nadif, Mohamed
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2017, 29 (01) : 2 - 16
  • [10] Classification and Clustering in Metagenomics with Unified Data Management and Computational Framework
    Rasheed, Zeehasham
    Rangwala, Huzefa
    Gillevet, Patrick
    2012 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE WORKSHOPS (BIBMW), 2012,