University of Sussex
Browse

File(s) not publicly available

Efficiency in unification-based n-best parsing

presentation
posted on 2023-06-08, 07:07 authored by Yi Zhang, Stephan Oepen, John Carroll
We extend a recently proposed algorithm for n-best unpacking of parse forests to deal efficiently with (a) Maximum Entropy (ME) parse selection models containing important classes of non-local features, and (b) forests produced by unification grammars containing significant proportions of globally inconsistent analyses. The new algorithm empirically exhibits a linear relationship between processing time and the number of analyses unpacked at all degrees of ME feature non-locality; in addition, compared with agenda-driven best-first parsing and exhaustive parsing with post-hoc parse selection it leads to improved parsing speed, coverage, and accuracy.

History

Publication status

  • Published

Publisher

Association for Computational Linguistics

Page range

48-59

Pages

12.0

Presentation Type

  • paper

Event name

Tenth International Conference on Parsing Technologies

Event location

Prague, Czech Republic

Event type

conference

ISBN

978-1-932432-90-9

Department affiliated with

  • Informatics Publications

Full text available

  • No

Peer reviewed?

  • Yes

Legacy Posted Date

2012-02-06

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC