Title
Constrained Locally Weighted Clustering
Abstract
Data clustering is a difficult problem due to the complex and heterogeneous natures of multidimensional data. To improve clustering accuracy, we propose a scheme to cap-ture the local correlation structures: associate each cluster with an independent weighting vector and embed it in the subspace spanned by an adaptive combination of the di-mensions. Our clustering algorithm takes advantage of the known pairwise instance-level constraints. The data points in the constraint set are divided into groups through in-ference; and each group is assigned to the feasible cluster which minimizes the sum of squared distances between all the points in the group and the corresponding centroid. Our theoretical analysis shows that the probability of points be-ing assigned to the correct clusters is much higher by the new algorithm, compared to the conventional methods. This is confirmed by our experimental results, indicating that our design indeed produces clusters which are closer to the ground truth than clusters created by the current state-of-the-art algorithms. © 2008 VLDB Endowment.
Publication Date
1-1-2008
Publication Title
Proceedings of the VLDB Endowment
Volume
1
Issue
1
Number of Pages
90-101
Document Type
Article
Personal Identifier
scopus
DOI Link
https://doi.org/10.14778/1453856.1453871
Copyright Status
Unknown
Socpus ID
77952760641 (Scopus)
Source API URL
https://api.elsevier.com/content/abstract/scopus_id/77952760641
STARS Citation
Cheng, Hao; Hua, Kien A.; and Vu, Khanh, "Constrained Locally Weighted Clustering" (2008). Scopus Export 2000s. 10584.
https://stars.library.ucf.edu/scopus2000/10584