GUPEA >
IT Faculty / IT-fakulteten >
Department of Computer Science and Engineering / Institutionen för data- och informationsteknik >
Doctoral Theses / Doktorsavhandlingar Institutionen för data- och informationsteknik >

Multilingual Abstractions: Abstract Syntax Trees and Universal Dependencies


Please use this identifier to cite or link to this item: http://hdl.handle.net/2077/60331

Files in This Item:

File Description SizeFormat
gupea_2077_60331_1.pdf3900KbAdobe PDF
View/Open
gupea_2077_60331_2.pdfThe spikblad222KbAdobe PDF
View/Open
Title: Multilingual Abstractions: Abstract Syntax Trees and Universal Dependencies
Authors: Kolachina, Prasanth
E-mail: prasanth.kolachina@cse.gu.se
Issue Date: 24-May-2019
University: Göteborgs universitet. IT-fakulteten
Institution: Department of Computer Science and Engineering ; Institutionen för data- och informationsteknik
Parts of work: Prasanth Kolachina and Aarne Ranta. "From Abstract Syntax to UniversalDependencies". Linguistic Issues in Language Technology 13(3), 2016

Aarne Ranta and Prasanth Kolachina "From Universal Dependencies to AbstractSyntax". Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies(UDW 2017), pp. 107–116.

Prasanth Kolachina and Martin Riedl and Chris Biemann. "Replacing OOV Words For Dependency Parsing With Distributional Semantics". Proceedings of the 21st Nordic Conference on Computational Linguistics (NoDaL-iDa), 2017, pp. 11–19.

Prasanth Kolachina and Aarne Ranta. Bootstrapping UD treebanks for Delexicalized Parsing. Under Submission.
Date of Defence: 2019-06-14
Disputation: June 14th 1000AM Room EA, Hörsalsvägen 11
Degree: Doctor of Philosophy
Publication type: Doctoral thesis
Series/Report no.: 174D
Keywords: Grammatical Framework
Universal Dependencies
Natural Language Processing
multilinguality
abstract syntax trees
dependency trees
multilingual generation
multilingual parsers
Abstract: This thesis studies the connections between parsing friendly representations and interlingua grammars developed for multilingual language generation. Parsing friendly representations refer to dependency tree representations that can be used for robust, accurate and scalable analysis of natural language text. Shared multilingual abstractions are central to both these representations. Universal Dependencies (UD) is a framework to develop cross-lingual representations, using dependency trees for mu... more
ISBN: 978-91-7833-509-1
URI: http://hdl.handle.net/2077/60331
Appears in Collections:Doctoral Theses from University of Gothenburg / Doktorsavhandlingar från Göteborgs universitet
Doctoral Theses / Doktorsavhandlingar Institutionen för data- och informationsteknik

 

 

© Göteborgs universitet 2011