A fast homotopy algorithm for a large class of weighted classification problems
Résumé
We propose a penalized method to solve the one-way ANOVA problem by collapsing the coefficients of K conditions. We introduce a large class of weights for which our homotopy algorithm is in O(K log(K)). These weights induce a balanced tree structure and simplify the interpretation of the results. As an example we consider phenotypic data: given a trait, we reconstruct a balanced tree structure and assess its agreement with the known phylogeny. Our proposal is easily extended to more than one dimension for clustering problems.
Domaines
Sciences du Vivant [q-bio]
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...