Show simple item record

dc.contributor.advisorMitkov, Ruslan
dc.contributor.advisorHa, Le An
dc.contributor.advisorYaneva, Victoria
dc.contributor.authorRohanian, Omid
dc.date.accessioned2020-12-18T14:49:06Z
dc.date.available2020-12-18T14:49:06Z
dc.date.issued2020
dc.identifier.urihttp://hdl.handle.net/2436/623843
dc.descriptionA thesis submitted in partial ful lment of the requirements of the University of Wolverhampton for the degree of Doctor of Philosophy.en
dc.description.abstractNon-literal language concerns the deliberate use of language in such a way that meaning cannot be inferred through a mere literal interpretation. In this thesis, three different forms of this phenomenon are studied; namely, irony, non-compositional Multiword Expressions (MWEs), and metaphor. We start by developing models to identify ironic comments in the context of the social micro-blogging website Twitter. In these experiments, we proposed a new way to extract features based on a study of their spatial structure. The proposed model is shown to perform competitively on a standard Twitter dataset. Next, we extensively study MWEs, which are the central point of focus in this work. We start by framing the task of MWE identi fication as sequence labelling and devise experiments to see the effect of eye-tracking data in capturing formulaic MWEs using structured prediction. We also develop a novel neural architecture to speci fically address the issue of discontinuous MWEs using a combination of Graph Convolutional Neural Networks (GCNs) and self-attention. The proposed model is subsequently tested on several languages where it is shown to outperform the state-of-the-art in overall criteria and also in capturing gappy MWEs. In the final part of the thesis, we look at metaphor and its interaction with verbal MWEs. In a series of experiments, we propose a hybrid BERT-based model augmented with a novel variation of GCN where we perform classifi cation on two standard metaphor datasets using information from MWEs. This model which performs at the same level with state-of-the-art is, to the best of our knowledge, the first MWE-aware metaphor identifi cation system paving the way for further experimentation on the interaction of different types of fi gurative language.en
dc.description.sponsorshipResearch Group in Computational Linguistics.en
dc.formatapplication/pdfen
dc.language.isoenen
dc.publisherUniversity of Wolverhamptonen
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 International*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectmetaphor processingen
dc.subjectirony detectionen
dc.subjectmultiword expressionsen
dc.subjectgraph convolutional networksen
dc.subjectdeep learningen
dc.subjecteye trackingen
dc.subjectsarcasmen
dc.subjectidiomsen
dc.subjectneural networksen
dc.titleContributions to the Computational Treatment of Non-literal Languageen
dc.typeThesis or dissertationen
dc.type.qualificationnamePhD
dc.type.qualificationlevelDoctoral
refterms.dateFOA2020-12-18T14:49:07Z


Files in this item

Thumbnail
Name:
Rohanian_PhD_thesis.pdf
Size:
2.109Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record

Attribution-NonCommercial-NoDerivatives 4.0 International
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivatives 4.0 International