Making senses: Bootstrapping sense-tagged lists of semantically-related words

Abstract

The work described in this paper was originally motivated by the need to map verbs associated with FrameNet 2.0 frames to appropriate WordNet 2.0 senses. As the work evolved, it became apparent that the developed method was applicable for a number of other tasks, including assignment of WordNet senses to word lists used in attitude and opinion analysis, and collapsing WordNet senses into coarser-grained groupings, We describe the method for mapping FrameNet lexical units to WordNet senses and demonstrate its applicability to these additional tasks. We conclude with a general discussion of the viability of using this method with automatically sense-tagged data. © Springer-Verlag Berlin Heidelberg 2006.