Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 22, 2026Digital humanities quarterlyOpen Access

Problems of Authorship Classification: Recognising the Author Style or a Book?

View Full Paper
Ask AI
Bookmark
Share

Authors

FVFrantišek VálekJHJr. Jan Hajič

Discussion

Loading...

Member takes

Overview

The article explores authorship attribution problems in a dataset of Czech authors, suggesting a bias towards book recognition.

Key Points

  • The aim is to investigate the issues in authorship attribution, focusing on distinguishing between individual books and overall author style.
  • Developed a diverse dataset of Czech authors for analysis
  • Utilized LinearSVC classifiers in a three-step experimental setup
  • Conducted attribution tasks with varying splits of training and testing data by book
  • Findings show authorship attribution tasks favor specific books over overall authors
  • Poorer performance is observed when books are split between training and testing
  • Authors with more internal confusion in their works exhibit better attribution performance for unseen books

Cite This Study

Válek et al. (2024) studied this question.

synapsesocial.com/papers/69bf8978f665edcd009e9260https://doi.org/10.63744/vd6r7u2cv5y4
View Full Paper
Ask AI
Bookmark
Share