An iterative optimization clustering algorithm based on manifold distance
Na Wang, Sun’an Wang, Haifeng Du · 2009
In this study, a novel iterative optimization clustering algorithm is proposed by using a manifold distance based dissimilarity metric which can measure the geodesic distance along the manifold and a criterion function which can express the clustering target, that is the samples in the same cluster being somehow more similar than samples in different one. The steps of the algorithm are discussed in detail. Simulation results on six artificial datasets with different manifold structures show that comparing to k means clustering algorithm, the new algorithm has the ability to identify complex non-convex clusters.