Logo: to the web site of Uppsala University

uu.sePublications from Uppsala University
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Temporally Classifying a Century of Swedish Government Official Reports Using Explainable AI
Uppsala University, Disciplinary Domain of Humanities and Social Sciences, Faculty of Social Sciences, Department of Statistics.
2022 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Publicly available Swedish Government Official Reports (Statens offentliga utredningar) released during the last 100 years are used for temporal classification according to their year of publication. Two BERT-based models are used to investigate how well a state-of-the-art NLPmodel can distinguish between Swedish documents based on the time they were written. The best-performing model is evaluated using two explainable artificial intelligence methods, in order to also gain an understanding of the logic behind the classifications. The findings show that it is possible to accurately separate and classify the reports according to their time of publication. The results from the explainability techniques show some inconsistencies, making it difficult to draw definitive conclusions about what words are most influential to the classifications.

Place, publisher, year, edition, pages
2022. , p. 54
Keywords [en]
temporal text classification, natural language processing, deep learning, explainable AI, BERT, Transformers
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:uu:diva-477114OAI: oai:DiVA.org:uu-477114DiVA, id: diva2:1669590
Educational program
Master Programme in Statistics
Supervisors
Examiners
Available from: 2022-06-17 Created: 2022-06-14 Last updated: 2022-06-17Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Department of Statistics
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 319 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf