Export Ready — 

Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data

Bibliographic Details
Main Author: Geller, Marcel
Publication Date: 2023
Format: Master thesis
Language: eng
Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
Download full: http://hdl.handle.net/10362/159459
Summary: Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
id RCAP_3e765fc62dbe05cc19ca60c803afe03a
oai_identifier_str oai:run.unl.pt:10362/159459
network_acronym_str RCAP
network_name_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository_id_str https://opendoar.ac.uk/repository/7160
spelling Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter dataTwitter ResearchTopic ModellingTopic EvolutionDiscourse ToxicityDynamic Topic ModellingSDG 16 - Peace, justice and strong institutionsDomínio/Área Científica::Ciências Naturais::Ciências da Computação e da InformaçãoDissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data ScienceThis thesis presents an extensive investigation into the evolution of topics and their association with speech toxicity on Twitter, based on a large corpus of tweets, providing crucial insights for monitoring online discourse and potentially informing interventions to combat toxic behavior in digital communities. A Dynamic Topic Evolution Model (DyTEM) is introduced, constructed by combining static Topic Modelling techniques and sentence embeddings through the state-of-the-art sentence transformer, sBERT. The DyTEM, tested and validated on a substantial sample of tweets, is represented as a directed graph, encapsulating the inherent dynamism of Twitter discussions. For validating the consistency of DyTEM and providing guidance for hyperparameter selection, a novel, hashtag-based validation method is proposed. The analysis identifies and scrutinizes five distinct Topic Transition Types: Topic Stagnation, Topic Merge, Topic Split, Topic Disappearance, and Topic Emergence. A speech toxicity classification model is employed to delve into the toxicity dynamics within topic evolution. A standout finding of this study is the positive correlation between topic popularity and its toxicity, implying that trending or viral topics tend to contain more inflammatory speech. This insight, along with the methodologies introduced in this study, contributes significantly to the broader understanding of digital discourse dynamics and could guide future strategies aimed at fostering healthier and more constructive online spaces.Pinheiro, Flávio Luís PortasRUNGeller, Marcel2023-11-02T14:22:48Z2023-10-232023-10-23T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttp://hdl.handle.net/10362/159459TID:203377540enginfo:eu-repo/semantics/openAccessreponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiainstacron:RCAAP2024-05-22T18:15:19Zoai:run.unl.pt:10362/159459Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireinfo@rcaap.ptopendoar:https://opendoar.ac.uk/repository/71602025-05-28T17:45:45.030128Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologiafalse
dc.title.none.fl_str_mv Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
title Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
spellingShingle Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
Geller, Marcel
Twitter Research
Topic Modelling
Topic Evolution
Discourse Toxicity
Dynamic Topic Modelling
SDG 16 - Peace, justice and strong institutions
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
title_short Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
title_full Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
title_fullStr Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
title_full_unstemmed Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
title_sort Toxicity in Evolving Twitter Topics - Employing a novel Dynamic Topic volution Model (DyTEM) onTwitter data
author Geller, Marcel
author_facet Geller, Marcel
author_role author
dc.contributor.none.fl_str_mv Pinheiro, Flávio Luís Portas
RUN
dc.contributor.author.fl_str_mv Geller, Marcel
dc.subject.por.fl_str_mv Twitter Research
Topic Modelling
Topic Evolution
Discourse Toxicity
Dynamic Topic Modelling
SDG 16 - Peace, justice and strong institutions
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
topic Twitter Research
Topic Modelling
Topic Evolution
Discourse Toxicity
Dynamic Topic Modelling
SDG 16 - Peace, justice and strong institutions
Domínio/Área Científica::Ciências Naturais::Ciências da Computação e da Informação
description Dissertation presented as the partial requirement for obtaining a Master's degree in Data Science and Advanced Analytics, specialization in Data Science
publishDate 2023
dc.date.none.fl_str_mv 2023-11-02T14:22:48Z
2023-10-23
2023-10-23T00:00:00Z
dc.type.status.fl_str_mv info:eu-repo/semantics/publishedVersion
dc.type.driver.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
status_str publishedVersion
dc.identifier.uri.fl_str_mv http://hdl.handle.net/10362/159459
TID:203377540
url http://hdl.handle.net/10362/159459
identifier_str_mv TID:203377540
dc.language.iso.fl_str_mv eng
language eng
dc.rights.driver.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.source.none.fl_str_mv reponame:Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
instname:FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron:RCAAP
instname_str FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
instacron_str RCAAP
institution RCAAP
reponame_str Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
collection Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
repository.name.fl_str_mv Repositórios Científicos de Acesso Aberto de Portugal (RCAAP) - FCCN, serviços digitais da FCT – Fundação para a Ciência e a Tecnologia
repository.mail.fl_str_mv info@rcaap.pt
_version_ 1833596946328059904

Similar Items