Keywords: kernel density estimation, k-means algorithm, overlap,syncytial
Syncytial clustering is the name that we introduce for methods that merge groups obtained from standard clustering algorithms. These algorithms usually find regular-structured clusters such as ellipsoidally- or spherically-dispersed groups, but are more challenged with groups lacking formal structure or definition. Here, we develop a distribution-free fully-automated syncytial clustering algorithm that can be used with k-means and other clustering algorithms. Our approach computes the cumulative distribution function of the normed residuals from an appropriately fit of k-groups model and calculates the nonparametric overlap between each pair of groups. Groups with high pairwise overlaps are merged together as long as the generalized overlap decreases or it approach the maximum overlap. Our methodology is always a top performer in identifying groups with regular and irregular structures in several datasets. Also our approach is used to identify the distinct kinds of activated regions in a functional Magnetic Resonance Imaging study.