Next Article in Journal
Interfacial Properties of Active-Passive Polymer Mixtures
Previous Article in Journal
Topographic Reconfiguration of Local and Shared Information in Anesthetic-Induced Unconsciousness

Projected Affinity Values for Nyström Spectral Clustering

Department of Electromechanical Engineering, Guangdong University of Technology, Guangzhou 510006, China
Department of Computing Science, University of Alberta, Edmonton, AB T6G 2R3, Canada
School of Automation, Northwestern Polytechnical University, Xi’an 710072, China
Author to whom correspondence should be addressed.
Entropy 2018, 20(7), 519;
Received: 19 May 2018 / Revised: 6 July 2018 / Accepted: 9 July 2018 / Published: 10 July 2018
(This article belongs to the Section Information Theory, Probability and Statistics)
In kernel methods, Nyström approximation is a popular way of calculating out-of-sample extensions and can be further applied to large-scale data clustering and classification tasks. Given a new data point, Nyström employs its empirical affinity vector, k, for calculation. This vector is assumed to be a proper measurement of the similarity between the new point and the training set. In this paper, we suggest replacing the affinity vector by its projections on the leading eigenvectors learned from the training set, i.e., using k*=i=1ckTuiui instead, where ui is the i-th eigenvector of the training set and c is the number of eigenvectors used, which is typically equal to the number of classes designed by users. Our work is motivated by the constraints that in kernel space, the kernel-mapped new point should (a) also lie on the unit sphere defined by the Gaussian kernel and (b) generate training set affinity values close to k. These two constraints define a Quadratic Optimization Over a Sphere (QOOS) problem. In this paper, we prove that the projection on the leading eigenvectors, rather than the original affinity vector, is the solution to the QOOS problem. The experimental results show that the proposed replacement of k by k* slightly improves the performance of the Nyström approximation. Compared with other affinity matrix modification methods, our k* obtains comparable or higher clustering performance in terms of accuracy and Normalized Mutual Information (NMI). View Full-Text
Keywords: Nyström approximation; out-of-sample; empirical affinity; machine learning Nyström approximation; out-of-sample; empirical affinity; machine learning
Show Figures

Figure 1

MDPI and ACS Style

He, L.; Zhu, H.; Zhang, T.; Yang, H.; Guan, Y. Projected Affinity Values for Nyström Spectral Clustering. Entropy 2018, 20, 519.

AMA Style

He L, Zhu H, Zhang T, Yang H, Guan Y. Projected Affinity Values for Nyström Spectral Clustering. Entropy. 2018; 20(7):519.

Chicago/Turabian Style

He, Li, Haifei Zhu, Tao Zhang, Honghong Yang, and Yisheng Guan. 2018. "Projected Affinity Values for Nyström Spectral Clustering" Entropy 20, no. 7: 519.

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

Back to TopTop