Learning Partial Lexicographic Preference Trees and Forests over Multi-Valued Attributes

Partial lexicographic preference trees, or PLP-trees, form an intuitive formalism for compact representation of qualitative preferences over combinatorial domains. We show that PLP-trees can be used to accurately model preferences arising in practical situations, and that high-accuracy PLP-trees can be effectively learned. We also propose and study learning methods for a variant of our model based on the concept of a PLP-forest, a collection of PLP-trees, where the preference order specified by a PLP-forest is obtained by aggregating the orders of its constituent PLP-trees. Our results demonstrate the potential of both approaches, with learning PLP-forests showing particularly promising behavior.