Generalizing Semantic Role Annotations Across Syntactically Similar Verbs (bibtex)
by Gordon, Andrew S. and Swanson, Reid
Abstract:
Large corpora of parsed sentences with semantic role labels (e.g. PropBank) provide training data for use in the creation of high-performance automatic semantic role labeling systems. Despite the size of these corpora, individual verbs (or role-sets) often have only a handful of instances in these corpora, and only a fraction of English verbs have even a single annotation. In this paper, we describe an approach for dealing with this sparse data problem, enabling accurate semantic role labeling for novel verbs (rolesets) with only a single training example. Our approach involves the identification of syntactically similar verbs found in PropBank, the alignment of arguments in their corresponding rolesets, and the use of their corresponding annotations in PropBank as surrogate training data.
Reference:
Generalizing Semantic Role Annotations Across Syntactically Similar Verbs (Gordon, Andrew S. and Swanson, Reid), In Proceedings of the 2007 Meeting of the Association for Computational Linguistics (ACL-07), 2007.
Bibtex Entry:
@inproceedings{gordon_generalizing_2007,
	address = {Prague, Czech Republic},
	title = {Generalizing {Semantic} {Role} {Annotations} {Across} {Syntactically} {Similar} {Verbs}},
	url = {http://ict.usc.edu/pubs/Generalizing%20Semantic%20Role%20Annotations%20Across%20Syntactically%20Similar%20Verbs.pdf},
	abstract = {Large corpora of parsed sentences with semantic role labels (e.g. PropBank) provide training data for use in the creation of high-performance automatic semantic role labeling systems. Despite the size of these corpora, individual verbs (or role-sets) often have only a handful of instances in these corpora, and only a fraction of English verbs have even a single annotation. In this paper, we describe an approach for dealing with this sparse data problem, enabling accurate semantic role labeling for novel verbs (rolesets) with only a single training example. Our approach involves the identification of syntactically similar verbs found in PropBank, the alignment of arguments in their corresponding rolesets, and the use of their corresponding annotations in PropBank as surrogate training data.},
	booktitle = {Proceedings of the 2007 {Meeting} of the {Association} for {Computational} {Linguistics} ({ACL}-07)},
	author = {Gordon, Andrew S. and Swanson, Reid},
	month = jun,
	year = {2007},
	keywords = {The Narrative Group}
}
Powered by bibtexbrowser