Can Semantic Roles Generalize Across Genres
作者： Szu-Ting Yi , Edward Loper , Martha Palmer , None
DOI:
关键词:
摘要: PropBank has been widely used as training data for Semantic Role Labeling. However, because this is taken from the WSJ, resulting machine learning models tend to overfit on idiosyncrasies of that text’s style, and do not port well other genres. In addition, since was designed a verb-by-verb basis, argument labels Arg2 - Arg5 get very diverse roles with inconsistent instances. For example, verb “make” uses “Material” argument; but “multiply” “Extent” argument. As result, it can be difficult automatic classifiers learn distinguish arguments Arg2-Arg5. We have created mapping between VerbNet provides thematic role label each verb-specific label. Since are more consistent across verbs, we able demonstrate these new easier learn.
aclweb.org参考文章(14)
Vasin Punyakanok, Wen-tau Yih, Dan Roth, The necessity of syntactic parsing for semantic role labeling international joint conference on artificial intelligence. pp. 1117- 1123 ,(2005)
Martha S. Palmer, Karin Kipper Schuler, Verbnet: a broad-coverage, comprehensive verb lexicon University of Pennsylvania. ,(2005)
Daniel Gildea, Julia Hockenmaier, Identifying semantic roles using Combinatory Categorial Grammar Proceedings of the 2003 conference on Empirical methods in natural language processing -. pp. 57- 64 ,(2003) , 10.3115/1119355.1119363
Beth Levin, English Verb Classes and Alternations: A Preliminary Investigation ,(1993)
Szu-ting Yi, Martha Palmer, None, The Integration of Syntactic Parsing and Semantic Role Labeling conference on computational natural language learning. pp. 237- 240 ,(2005) , 10.3115/1706543.1706590
David Dowty, Thematic proto-roles and argument selection Language. ,vol. 67, pp. 547- 619 ,(1991) , 10.1353/LAN.1991.0021
John Chen, Owen Rambow, Use of deep linguistic features for the recognition and labeling of semantic arguments Proceedings of the 2003 conference on Empirical methods in natural language processing -. pp. 41- 48 ,(2003) , 10.3115/1119355.1119361
Mitchell Marcus, Grace Kim, Mary Ann Marcinkiewicz, Robert MacIntyre, Ann Bies, Mark Ferguson, Karen Katz, Britta Schasberger, The Penn Treebank Proceedings of the workshop on Human Language Technology - HLT '94. pp. 114- 119 ,(1994) , 10.3115/1075812.1075835
Sameer Pradhan, Wayne Ward, Kadri Hacioglu, James H. Martin, Daniel Jurafsky, Semantic Role Labeling Using Different Syntactic Views meeting of the association for computational linguistics. pp. 581- 588 ,(2005) , 10.3115/1219840.1219912
Xavier Carreras, Lluís Màrquez, Introduction to the CoNLL-2005 Shared Task: Semantic Role Labeling conference on computational natural language learning. pp. 152- 164 ,(2005) , 10.3115/1706543.1706571