Simplicial clustering using the $α$--transformation
By: Michail Tsagris, Nikolaos Kontemeniotis
Potential Business Impact:
Groups similar data points together more accurately.
We introduce two simplicial clustering approaches for compositional data, that are adaptations of the $K$--means and of the Gaussian mixture models algorithms, by employing the $\alpha$--transformation. By utilizing clustering validation indices we can decide on the number of clusters and choose the value of $\alpha$ for the $K$--means, while for the model-based clustering approach information criteria complete this task. extensive simulation studies compare the performance of these two approaches and a real data set illustrates their performance in real world settings.
Similar Papers
Simplicial clustering using the $α$--transformation
Methodology
Groups similar data points together more accurately.
The $α$--regression for compositional data: a unified framework for standard, spatially-lagged, and geographically-weighted regression models
Methodology
Helps scientists understand how parts make up a whole.
Comprehensive Analysis and Exclusion Hypothesis of $α$-Approximation Method for Discretizing Analog Systems
Systems and Control
Makes computer models work better with real-world signals.