DocumentCode
2352348
Title
Exploring Various Features in Semantic Role Labeling
Author
Hongling Wang ; Guodong Zhou ; Qiaoming Zhu ; Peide Qian
Author_Institution
Sch. of Comput. Sci. & Technol., Soochow Univ., Suzhou
fYear
2008
fDate
23-25 July 2008
Firstpage
3
Lastpage
8
Abstract
This paper explores the contributions of various features in semantic role labeling. Moreover, an optimal set of features is selected using a greedy strategy. Finally, an effective headword-driven pruning algorithm is proposed to filter out irrelavant instances. Evaluation on the CoNLL´2005 SRL benchmark corpus shows that our method achieved comparable performance with the best-reported ones on a single automatic parse tree.
Keywords
grammars; greedy algorithms; programming language semantics; greedy strategy; headword-driven pruning algorithm; semantic role labeling; single automatic parse tree; Computer science; Filters; Gold; Information processing; Information technology; Labeling; Magnetic heads; Support vector machine classification; Support vector machines; System performance; Semantic Role Labeling;
fLanguage
English
Publisher
ieee
Conference_Titel
Advanced Language Processing and Web Information Technology, 2008. ALPIT '08. International Conference on
Conference_Location
Dalian Liaoning
Print_ISBN
978-0-7695-3273-8
Type
conf
DOI
10.1109/ALPIT.2008.74
Filename
4584332
Link To Document