Change search
ReferencesLink to record
Permanent link

Direct link
Learning Structural Kernels for Natural Language Processing
University of Sheffield.
University of Melbourne.
Uppsala University, Disciplinary Domain of Humanities and Social Sciences, Faculty of Languages, Department of Linguistics and Philology. (Datorlingvistik)
University of Sheffield.
2015 (English)In: Transactions of the Association for Computational Linguistics, ISSN 2307-387X, Vol. 3, 461-473 p.Article in journal (Refereed) Published
Abstract [en]

Structural kernels are a flexible learning paradigm that has been widely used in Natural Language Processing. However, the problem of model selection in kernel-based methods is usually overlooked. Previous approaches mostly rely on setting default values for kernel hyperparameters or using grid search, which is slow and coarse-grained. In contrast, Bayesian methods allow efficient model selection by maximizing the evidence on the training data through gradient-based methods. In this paper we show how to perform this in the context of structural kernels by using Gaussian Processes. Experimental results on tree kernels show that this procedure results in better prediction performance compared to hyperparameter optimization via grid search. The framework proposed in this paper can be adapted to other structures besides trees, e.g., strings and graphs, thereby extending the utility of kernel-based methods.

Place, publisher, year, edition, pages
Stroudsburg, PA: Association for Computational Linguistics, 2015. Vol. 3, 461-473 p.
National Category
Language Technology (Computational Linguistics) Other Computer and Information Science
Research subject
Computational Linguistics
URN: urn:nbn:se:uu:diva-260587OAI: diva2:847771
Available from: 2015-08-21 Created: 2015-08-21 Last updated: 2015-09-07Bibliographically approved

Open Access in DiVA

TACL2015(358 kB)46 downloads
File information
File name FULLTEXT01.pdfFile size 358 kBChecksum SHA-512
Type fulltextMimetype application/pdf

Other links

Search in DiVA

By author/editor
Hardmeier, Christian
By organisation
Department of Linguistics and Philology
Language Technology (Computational Linguistics)Other Computer and Information Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 46 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 265 hits
ReferencesLink to record
Permanent link

Direct link