Spectral Rotation versus K-Means in Spectral Clustering
Year: 2013
Type: conference-paper
Abstract: Spectral clustering has been a popular data clustering algorithm. This category of approaches often resort to other clustering methods, such as K-Means, to get the final cluster. The potential flaw of such common practice is that the obtained relaxed continuous spectral solution could severely deviate from the true discrete solution. In this paper, we propose to impose an additional orthonormal constraint to better approximate the optimal continuous solution to the graph cut objective functions. Such a method, called spectral rotation in literature, optimizes the spectral clustering objective functions better than K-Means, and improves the clustering accuracy. We would provide efficient algorithm to solve the new problem rigorously, which is not significantly more costly than K-Means. We also establish the connection between our method andK-Means to provide theoretical motivation of our method. Experimental results show that our algorithm consistently reaches better cut and meanwhile outperforms in clustering metrics than classic spectral clustering methods.
Authors: Jin Huang, Feiping Nie, Heng Huang
Institution: The University of Texas at Arlington
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Funder: National Science Foundation
Awards: Collaborative Research: Non-negative Matrix Factorizations for Data Mining: Foundations, Capabilities, and Applications, Collaborative Research: Matrix-Model Machine Learning: Unifying Machine Learning and Scientific Computing, III: Small: Collaborative Research: A Large-Scale Data Mining Framework for Genome-Wide Mapping of Multi-Modal Phenotypic Biomarkers and Outcome Prediction, New Theoretical Foundations of Tensor Applications: Clustering, Error Analysis, Global Convergence, and Robust Formulations
OpenAlex ID: W62192748