Show simple item record

Files in this item

Thumbnail
Name:
fsmnlp2.pdf
Size:
213.0Kb
Format:
PDF
View/Open

Count-based state merging for probabilistic regular tree grammars

Item metadata

dc.contributor.authorDietze, Toni
dc.contributor.authorNederhof, Mark Jan
dc.date.accessioned2015-08-21T11:40:03Z
dc.date.available2015-08-21T11:40:03Z
dc.date.issued2015-06-22
dc.identifier.citationDietze , T & Nederhof , M J 2015 , Count-based state merging for probabilistic regular tree grammars . in Proceedings of the 12th International Conference on Finite State Methods and Natural Language Processing . Association for Computational Linguistics , Duesseldorf, Germany , The 12th International Conference on Finite-State Methods and Natural Language Processing , Düsseldorf , Germany , 22/06/15 .en
dc.identifier.citationconferenceen
dc.identifier.otherPURE: 198221584
dc.identifier.otherPURE UUID: e5f16c06-1097-4ba0-93a4-d6ab7624fdf7
dc.identifier.otherORCID: /0000-0002-1845-6829/work/46002721
dc.identifier.otherScopus: 85119663701
dc.identifier.urihttps://hdl.handle.net/10023/7301
dc.description.abstractWe present an approach to obtain language models from a tree corpus using probabilistic regular tree grammars (prtg). Starting with a prtg only generating trees from the corpus, the prtg is generalized step by step by merging nonterminals. We focus on bottom-up deterministic prtg to simplify the calculations.
dc.format.extent7
dc.language.isoeng
dc.publisherAssociation for Computational Linguistics
dc.relation.ispartofProceedings of the 12th International Conference on Finite State Methods and Natural Language Processingen
dc.rightsCopyright ACL 2015. Licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License.en
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subjectT-NDASen
dc.subject.lccQA75en
dc.titleCount-based state merging for probabilistic regular tree grammarsen
dc.typeConference itemen
dc.description.versionPostprinten
dc.contributor.institutionUniversity of St Andrews. School of Computer Scienceen
dc.identifier.urlhttp://fsmnlp2015.phil.hhu.de/en


This item appears in the following Collection(s)

Show simple item record

Items in the St Andrews Research Repository are protected by copyright, with all rights reserved, unless otherwise indicated.