Modelling Speaker Attribution in Narrative Texts With Biased and Bias-Adjustable Neural Networks
2022-02-03 | journal article; research paper. A publication with affiliation to the University of Göttingen.
Jump to: Cite & Linked | Documents & Media | Details | Version history
Cite this publication
Modelling Speaker Attribution in Narrative Texts With Biased and Bias-Adjustable Neural Networks
Dönicke, T. ; Varachkina, H. ; Weimer, A. M.; Gödeke, L.; Barth, F.; Gittel, B. & Holler, A. et al. (2022)
Frontiers in Artificial Intelligence, 4 art. 725321. DOI: https://doi.org/10.3389/frai.2021.725321
Documents & Media
Data_Sheet_1.PDF169.79 kBAdobe PDFfrai-04-725321-g0001.tif222.65 kBTIFFfrai-04-725321-g0002.tif219.64 kBTIFFfrai-04-725321-i0001.tif204.91 kBTIFFfrai-04-725321.pdf635.17 kBAdobe PDF
Details
- Authors
- Dönicke, Tillmann ; Varachkina, Hanna ; Weimer, Anna Mareike; Gödeke, Luisa; Barth, Florian; Gittel, Benjamin ; Holler, Anke ; Sporleder, Caroline
- Abstract
- Literary narratives regularly contain passages that different readers attribute to different speakers: a character, the narrator, or the author. Since literary narratives are highly ambiguous constructs, it is often impossible to decide between diverging attributions of a specific passage by hermeneutic means. Instead, we hypothesise that attribution decisions are often influenced by annotator bias, in particular an annotator's literary preferences and beliefs. We present first results on the correlation between the literary attitudes of an annotator and their attribution choices. In a second set of experiments, we present a neural classifier that is capable of imitating individual annotators as well as a common-sense annotator, and reaches accuracies of up to 88% (which improves the majority baseline by 23%).
- Issue Date
- 3-February-2022
- Journal
- Frontiers in Artificial Intelligence
- Organization
- Seminar für Deutsche Philologie ; Göttingen Centre for Digital Humanities
- ISSN
- 2624-8212
- Language
- English