Given a sample of reference documents from a restricted and finite set of candidate authors, the task is to determine the most likely author of a previously unseen document of unknown authorship. Documents of known and unknown authorship belong to different domains (fandoms). More specifically, all documents of unknown authorship are fics of the same fandom (target fandom) while the documents of known authorship by the candidate authors are fics of several fandoms (other than the target-fandom). The participants are asked to prepare a method that can handle multiple cross-fandom attribution problems.
Publication
Mike Kestemont, Michael Tschuggnall, Efstathios Stamatatos, Walter Daelemans, Günther Specht, Benno Stein, and Martin Potthast (2018) Overview of the Author Identification Task at PAN-2018: Cross-domain Authorship Attribution and Style Change Detection. In Linda Cappellato, Nicola Ferro, Jian-Yun Nie, and Laure Soulier, editors, Working Notes Papers of the CLEF 2018 Evaluation Labs, volume 2125 of CEUR Workshop Proceedings, September 2018. CEUR-WS.org.
Competition
Language
Spanish
English
NLP topic
Abstract task
Dataset
Year
2018
Publication link
Ranking metric
Macro F1
Task results
| System | MacroF1 Sort ascending |
|---|---|
| Custdio and Paraboni | 0.8560 |
| Halvani and Graner | 0.7510 |
| Murauer et al. | 0.7340 |
| Mosavat | 0.7310 |
| Martn dCR et al. | 0.7050 |
