PCFG learning by nonterminal partition search

BELZ, ANJA (2002) PCFG learning by nonterminal partition search In: Grammatical Inference: Algorithms and Applications: Proceedings of the 6th International Colloquium: ICGI 2002, Amsterdam, The Netherlands, September 23-25, 2002 .

Restricted to Registered users only


Official URL: http://www.springerlink.com/content/hlld0q54m307/


pcfg Learning by Partition Search is a general grammatical inference method for constructing, adapting and optimising pcfgs. Given a training corpus of examples from a language, a canonical grammar for the training corpus, and a parsing task, Partition Search pcfg Learning constructs a grammar that maximises performance on the parsing task and minimises grammar size. This paper describes Partition Search in detail, also providing theoretical background and a characterisation of the family of inference methods it belongs to. The paper also reports an example application to the task of building grammars for noun phrase extraction, a task that is crucial in many applications involving natu- ral language processing. In the experiments, Partition Search improves parsing performance by up to 21.45% compared to a general baseline and by up to 3.48% compared to a task-specific baseline, while reducing grammar size by up to 17.25%.

Item Type:Contribution to conference proceedings in the public domain ( Full Paper)
Uncontrolled Keywords:Partition search
Subjects:Q000 Languages and Literature - Linguistics and related subjects > Q100 Linguistics
DOI (a stable link to the resource):
Faculties:Faculty of Science and Engineering > School of Computing, Engineering and Mathematics > Natural Language Technology
ID Code:3204
Deposited By:Converis
Deposited On:18 Nov 2007
Last Modified:22 Nov 2013 12:58

Repository Staff Only: item control page