Publishing Partner: Cambridge University Press CUP Extra Wiley-Blackwell Publisher Login
amazon logo
More Info


New from Oxford University Press!

ad

The Vulgar Tongue: Green's History of Slang

By Jonathon Green

A comprehensive history of slang in the English speaking world by its leading lexicographer.


New from Cambridge University Press!

ad

The Universal Structure of Categories: Towards a Formal Typology

By Martina Wiltschko

This book presents a new theory of grammatical categories - the Universal Spine Hypothesis - and reinforces generative notions of Universal Grammar while accommodating insights from linguistic typology.


New from Brill!

ad

Brill's MyBook Program

Do you have access to Dynamics of Morphological Productivity through your library? Then you can by the paperback for only €25 or $25! Find out more about Brill's MyBook program!


Academic Paper


Title: Using Machine-Learning to Assign Function Labels to Parser Output for Spanish
Paper URL: http://acl.ldc.upenn.edu/P/P06/P06-2018.pdf
Author: Grzegorz Chrupała
Email: click here to access email
Homepage: http://www.lsv.uni-saarland.de/personalPages/gchrupala/index.html
Institution: Saarland University
Author: Josef Van Genabith
Email: click here to access email
Institution: Dublin City University
Linguistic Field: Computational Linguistics
Subject Language: Spanish
Spanish
Abstract: Data-driven grammatical function tag assignment has been studied for English using the Penn-II Treebank data. In this paper we address the question of whether such methods can be applied successfully to other languages and treebank resources. In addition to tag assignment accuracy and f-scores we also present results of a task-based evaluation. We use three machine-learning methods to assign Cast3LB function tags to sentences parsed with Bikel's parser trained on the Cast3LB treebank. The best performing method, SVM, achieves an f-score of 86.87% on gold-standard trees and 66.67% on parser output - a statistically significant improvement of 6.74% over the baseline. In a task-based evaluation we generate LFG functional-structures from the function-tag-enriched trees. On this task we achieve an f-score of 75.67%, a statistically significant 3.4% improvement over the baseline.
Type: Individual Paper
Status: Completed
Publication Info: Proceedings of the COLING/ACL 2006 Main Conference Poster Sessions
URL: http://acl.ldc.upenn.edu/P/P06/P06-2018.pdf


Back
Add a new paper
Return to Academic Papers main page
Return to Directory of Linguists main page