Browse DORAS
Browse Theses
Search
Latest Additions
Creative Commons License
Except where otherwise noted, content on this site is licensed for use under a:

Structured parameter estimation for LFG-DOP using Backoff

Hearne, Mary and Sima'an, Khalil (2003) Structured parameter estimation for LFG-DOP using Backoff. In: RANLP 2003 - Recent Advances in Natural Language Processing, Borovets, Bulgaria, 10-12 September 2003.

Full text available as:

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
127Kb

Abstract

Despite its state-of-the-art performance, the Data Oriented Parsing (DOP) model has been shown to suffer from biased parameter estimation, and the good performance seems more the result of ad hoc adjustments than correct probabilistic generalization over the data. In recent work, we developed a new estimation procedure, called Backoff Estimation, for DOP models that are based on Phrase-Structure annotations (so called Tree-DOP models). Backoff Estimation deviates from earlier methods in that it treats the model parameters as a highly structured space of correlated events (backoffs), rather than a set of disjoint events. In this paper we show that the problem of biased estimates also holds for DOP models that are based on Lexical-Functional Grammar annotations (i.e. LFG-DOP), and that the LFG-DOP parameters also constitute a hierarchically structured space. Subsequently, we adapt the Backoff Estimation algorithm from Tree-DOP to LFG-DOP models. Backoff Estimation turns out to be a natural solution to some of the specific problems of robust parsing under LFGDOP.

Item Type:Conference or Workshop Item (Paper)
Event Type:Conference
Refereed:Yes
Uncontrolled Keywords:data oriented parsing (DOP) model;
Subjects:Computer Science > Machine translating
DCU Faculties and Centres:Research Initiatives and Centres > National Centre for Language Technology (NCLT)
DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing
Official URL:http://lml.bas.bg/ranlp2003/
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
ID Code:15315
Deposited On:16 Mar 2010 11:49 by DORAS Administrator. Last Modified 28 Apr 2010 11:59

Download statistics

Archive Staff Only: edit this record