TiK-means: $K$-means clustering for skewed groups
This is an incremental improvement for researchers in clustering and astronomy, enabling better analysis of skewed data distributions.
The authors tackled the problem of clustering skewed groups by extending the K-means algorithm to estimate skewness-transformation parameters, resulting in TiK-means, which was evaluated on simulated and real datasets and applied to resolve an astronomical dispute about gamma ray bursts.
The $K$-means algorithm is extended to allow for partitioning of skewed groups. Our algorithm is called TiK-Means and contributes a $K$-means type algorithm that assigns observations to groups while estimating their skewness-transformation parameters. The resulting groups and transformation reveal general-structured clusters that can be explained by inverting the estimated transformation. Further, a modification of the jump statistic chooses the number of groups. Our algorithm is evaluated on simulated and real-life datasets and then applied to a long-standing astronomical dispute regarding the distinct kinds of gamma ray bursts.