PCA-guided search for K-means

K-means is undoubtedly themostwidely used partitional clustering algorithm. Unfortunately, due to the nonconvexity of the model formulations, expectation-maximization (EM) type algorithms converge to different local optima with different initializations. Recent discoveries have identified that the global solution of K-means cluster centroids lies in the principal component analysis (PCA) subspace. Based on this insight, we propose PCA-guided effective search for K-means. Because the PCA subspace ismuch smaller than the original space, searching in the PCA subspace is both more effective and efficient. Extensive experiments on four real world data sets and systematic comparisonwith previous algorithms demonstrate that our proposedmethod outperforms the rest as it makes the K-means more effective. © 2015 Elsevier B.V. All rights reserved.