Density sensitive based spectral clustering

Abstract

The clustering effect by using spectral method depends heavily on the description of similarity between instances of the datasets. In this paper, we introduce a density sensitive distance measure which squeezes the distances in high density regions while widening them in low density regions. Experimental results show that compared with conventional spectral clustering algorithms, our proposed algorithm with density sensitive similarity measure can obtain desirable clusters with high performance. © 2011 Springer Science+Business Media B.V.