liu.seSearch for publications in DiVA
Endre søk
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Visual Analysis of Humor Assessment in Edited News Headlines
Linköpings universitet, Institutionen för teknik och naturvetenskap, Medie- och Informationsteknik. Linköpings universitet, Tekniska fakulteten.
Linköpings universitet, Institutionen för teknik och naturvetenskap, Medie- och Informationsteknik. Linköpings universitet, Tekniska fakulteten.
2023 (engelsk)Independent thesis Advanced level (degree of Master (One Year)), 20 poäng / 30 hpOppgave
Abstract [en]

In this project, we have focused on the Humicroedit data set which was created for one of the tasks presented during SemEval 2020. It contains the micro-edited versions of existing headlines where the aim was to make them funny. To analyze the edited headlines, various NLP techniques were applied, where the goal was to analyze the relationships and trends between the humor scores, sentiments, and topics of each data item. We have also considered the named entities and keywords, although to a lesser extent. This resulted in a visualization prototype that utilized the information visualization techniques treemap and dimensionality reduction. With the help of filtering and exploration features, an analysis could be performed on the data from the perspective of the different NLP modules. For this particular data set it was found that depending on the sentiment module, items of a certain score range and sentiment range will be grouped differently. It could also be determined that the sentiment value and the funniness score were highly dependent on the context of the edited headline. No certain connection could be made on how the topic affected the funniness score or the sentiment value due to the imbalanced distribution of topics in the Humicroedit data set. The prototype was evaluated by experts in NLP research and related fields. They deemed the prototype useful for its purpose and saw potential in exploring similar data sets with it, as well as reusing some of its features in their line of work.

sted, utgiver, år, opplag, sider
2023. , s. 95
Emneord [en]
Visual Text Analytics, Information Visualization, NLP
HSV kategori
Identifikatorer
URN: urn:nbn:se:liu:diva-199904ISRN: LiU-ITN-TEK-A--23/042--SEOAI: oai:DiVA.org:liu-199904DiVA, id: diva2:1823797
Fag / kurs
Media Technology
Uppsök
Technology
Veileder
Examiner
Merknad

Examensarbetet är utfört vid Institutionen för teknik och naturvetenskap (ITN) vid Tekniska fakulteten, Linköpings universitet

Tilgjengelig fra: 2024-01-03 Laget: 2024-01-03 Sist oppdatert: 2024-01-03bibliografisk kontrollert

Open Access i DiVA

Fulltekst mangler i DiVA

Søk i DiVA

Av forfatter/redaktør
Akkurt, ElinFolde, Johanna
Av organisasjonen

Søk utenfor DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric

urn-nbn
Totalt: 152 treff
RefereraExporteraLink to record
Permanent link

Direct link
Referera
Referensformat
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Annet format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annet språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf