Fast large-scale spectral clustering by sequential shrinkage optimization

Authors:
Tie-Yan Liu;Huai-Yuan Yang;Xin Zheng;Tao Qin;Wei-Ying Ma
Affiliations:
Microsoft Research Asia, Haidian District, Beijing, P.R. China;Peking University, Beijing, P.R. China;Tsinghua University, Beijing, P.R. China;Tsinghua University, Beijing, P.R. China;Microsoft Research Asia, Haidian District, Beijing, P.R. China
Venue:
ECIR'07 Proceedings of the 29th European conference on IR research
Year:
2007

Citing 4
Cited 3

Normalized Cuts and Image Segmentation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Toward the Optimal Preconditioned Eigensolver: Locally Optimal Block Preconditioned Conjugate Gradient Method

SIAM Journal on Scientific Computing
Preconditioned Eigensolvers: Practical Algorithms

Preconditioned Eigensolvers: Practical Algorithms
Multiclass Spectral Clustering

ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2

COACH: cumulative online algorithm for classification of handwriting deficiencies

IAAI'08 Proceedings of the 20th national conference on Innovative applications of artificial intelligence - Volume 3
p-PIC: Parallel power iteration clustering for big data

Journal of Parallel and Distributed Computing
Large-scale spectral clustering on graphs

IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

In many applications, we need to cluster large-scale data objects. However, some recently proposed clustering algorithms such as spectral clustering can hardly handle large-scale applications due to the complexity issue, although their effectiveness has been demonstrated in previous work. In this paper, we propose a fast solver for spectral clustering. In contrast to traditional spectral clustering algorithms that first solve an eigenvalue decomposition problem, and then employ a clustering heuristic to obtain labels for the data points, our new approach sequentially decides the labels of relatively well-separated data points. Because the scale of the problem shrinks quickly during this process, it can be much faster than the traditional methods. Experiments on both synthetic data and a large collection of product records show that our algorithm can achieve significant improvement in speed as compared to traditional spectral clustering algorithms.