Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • apa
  • ieee
  • modern-language-association-8th-edition
  • 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
Representing the aboutness of fiction: A comparison of manual and computational perspectives
University of Borås, Faculty of Librarianship, Information, Education and IT. (KIR)ORCID iD: 0000-0001-8184-8051
2023 (English)In: / [ed] Jack Andersen, University of Copenhagen and Joacim Hansson, Linnaeus University, Växjö, 2023Conference paper, Oral presentation with published abstract (Other academic)
Abstract [en]

The classification and indexing of literary fiction has proven a notoriously challenging area in LIS studies. Those engaging with this task often encounter a multitude of problems that, at times, make the endeavor seem close to unsolvable to a satisfactory degree. The intentional openness of fictional texts, and the recurring problem of how to representatively summarize their aboutness into sets of keywords (such as form/genre terms or subject headings from controlled vocabularies), pose a persistent challenge for knowledge organization theorists and professionals. This presentation attempts to approach the problem from a relatively unexplored angle; namely, that of employing contemporary, computational methods to assist in this task. The presentation departs from a short review of known problems in traditional fiction indexing theory and practice, and draws from actual examples and comparisons from existing library catalogues and organizational systems. In the presentation, examples of general-purpose fiction indexing from the Swedish national library catalogue LIBRIS are compared to indexing performed by Queerlit (a more recent initiative with the specific aim of indexing literary fiction from a LGBTQI perspective). The presentation then introduces the computational method of topic modeling as a means of revealing underlying topical patterns in unstructured fiction collections (in terms of genre and subject metadata). Although often employed by scholars aiming to reveal hidden topics in large collections of texts, topic modeling has seldom been used as a means of organizing fiction collections for library catalogues. Tentative findings from experiments examining the use of topic modeling on a collection of Swedish literary fiction are then discussed in relation to the indexing procedures applied in LIBRIS and Queerlit. The presentation concludes by discussing strengths and drawbacks of each of the three approaches, with particular emphasis on what aspects of fictional content is revealed and what is excluded in each different perspective.

Place, publisher, year, edition, pages
Växjö, 2023.
Keywords [en]
classification, indexing, fiction, literature, topic modeling, text mining
National Category
Information Studies
Research subject
Library and Information Science
Identifiers
URN: urn:nbn:se:hb:diva-31051OAI: oai:DiVA.org:hb-31051DiVA, id: diva2:1821466
Conference
Cultivating Information, Organizing Culture: Exploring Classification Practices in Modern Culture, Copenhagen, December 7-8, 2023.
Available from: 2023-12-20 Created: 2023-12-20 Last updated: 2024-01-10Bibliographically approved

Open Access in DiVA

No full text in DiVA

Search in DiVA

By author/editor
Falk, Olof
By organisation
Faculty of Librarianship, Information, Education and IT
Information Studies

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 61 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • apa
  • ieee
  • modern-language-association-8th-edition
  • 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