Projecting semantic roles via Tai mappings
Hector-Hugo Franco-Penya, Martin Emms; Proceedings of KONVENS 2012 (Main track: oral presentations), pp. 61-69, September 2012.
This work takes the paradigm of projecting annotations within labelled data into unlabelled data, via a mapping, and applies it to Semantic Role Labelling. The projections are amongst dependency trees and the mappings are the Tai-mappings that underlie the well known tree edit-distance algorithm. The system was evaluated in seven different languages. A number of variants are explored relating to the amount of information attended to in aligning nodes, whether the scoring is distance-based or similarity-based , and the relative ease with which nodes can be ignored. We find that all of these have statistically significant impacts on the outcomes, mostly in language-independent ways, but sometimes language dependently.