Is bad structure better than no structure?

unsupervised parsing for realisation ranking

Yasaman Motazedi*, Mark Dras, François Lareau

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

1 Citation (Scopus)

Abstract

In natural language generation using symbolic grammars, state-of-the-art realisation rankers use statistical models incorporating both language model and structural features. The rankers depend on multiple structures produced by the particular large-scale symbolic grammars to rank the output; for languages with smaller resources and in-development grammars, we look at the feasibility of an alternative source of structural features, unsupervised parsers. We show that, in spite of their lower quality of structure, raw sets of unsupervised parse features can be helpful with smaller language models; and that the parses do contain particular elements that can be highly useful, improving performance on our classification task by up to 10% on 60% of the test set leading to an overall improvement under a back-off model.

Original languageEnglish
Title of host publication24th International Conference on Computational Linguistics
Subtitle of host publicationProceedings of COLING 2012: Technical Papers
EditorsMartin Kay, Christian Boitet
Place of PublicationMumbai
PublisherIndian Institute of Technology
Pages1811-1830
Number of pages20
Publication statusPublished - 2012
Event24th International Conference on Computational Linguistics, COLING 2012 - Mumbai, India
Duration: 8 Dec 201215 Dec 2012

Other

Other24th International Conference on Computational Linguistics, COLING 2012
CountryIndia
CityMumbai
Period8/12/1215/12/12

Fingerprint Dive into the research topics of 'Is bad structure better than no structure? unsupervised parsing for realisation ranking'. Together they form a unique fingerprint.

Cite this