Output kernel tree
In this paper, we generalize tree-based methods to the prediction in structured output space. The extension is based on a kernelization of the algorithm that allows one to grow trees as soon as a kernel can be defined on the output space. The resulting algorithm, called output kernel trees (OK3), generalizes classification and regression trees in a principled way.
Author: Florence d'Alché, Université Evry Val d'Essonne Genopole