Mapping WordNets Using Structural Information

摘要

We present a robust approach for linking already existing lexical/semantichierarchies. We used a constraint satisfaction algorithm (relaxation labeling)to select --among a set of candidates-- the node in a target taxonomy thatbests matches each node in a source taxonomy. In particular, we use it to mapthe nominal part of WordNet 1.5 onto WordNet 1.6, with a very high precisionand a very low remaining ambiguity.