Augmenting Stylometric Features to Improve Detection of Propaganda and Manipulation

Warning

This publication doesn't include Faculty of Arts. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

SABOL Radoslav HORÁK Aleš

Year of publication 2023
Type Article in Proceedings
Conference Proceedings of the Seventeenth Workshop on Recent Advances in Slavonic Natural Languages Processing, RASLAN 2023
MU Faculty or unit

Faculty of Informatics

Citation
Web
Keywords stylometry; propaganda detection; manipulative style analysis; Propaganda dataset; Czech
Description Identification of manipulative techniques in newspaper texts allows an informed reader to cope with the text content without being negatively influenced. In this paper, we present new developments in using stylometry to support a deep learning neural network model in labelling newspaper articles for the presence of specific manipulative techniques. We also evaluate all stylometric features in 16 groups and improve the manipulation detection results in 15 of 17 techniques.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.