Maximum Entropy Markov Models for Semantic Role Labelling
Phil Blunsom · 2004
This paper investigates the application of Maximum Entropy Markov Models to semantic role labelling. Syntactic chunks are labelled according to the semantic role they fill for sentence verb predicates. The model is trained on the subset of Propbank data provided for the Conference on Computational Natural Language Learning 2004. Good precision is achieved, which is of key importance for information extraction from large corpora containing redundant data, and for generalising systems beyond task specific, hand coded template methods.