Abstract
Recent years have shown the power of dripping propaganda poison in minds of large populations with the goals to earn their support and to justify actions against particular groups or nations. These instruments of dripping the poison have been analysed in the literature on automated bias detection mainly from the lexical perspective. However, a more comprehensive perspective is needed for a more complete understanding and modelling of political biases present in this type of communication within such communities. To achieve our aim of a better perspective, we identified a case of such political propaganda that exerts influence and affects the general perception of one group by another one, which is general enough, well known enough but yet specific enough for our aim, namely Russia’s portraying Ukrainians as right radicals. We analysed the instruments of bias in 68 articles in 4 languages reporting an event of fire in the House of Trade Unions in Odessa (Ukraine) on May 2, 2014. The analysis methods are inspired by the concepts of event structure, linguistic analysis and emotion studies. We identified three main dimensions of bias: logic, lexicon and emotionality, which are constitutive of the analysed emotionally, lexically and propositionally biased reports. Moreover, we extracted 28 types of instruments of bias which use the three main dimensions of bias to a different extent. Lexical choices are just one of them. This research puts forward a basic coherent classification of the instruments of bias for further computational and formal modelling.
Original language | English |
---|---|
Pages (from-to) | 555-585 |
Number of pages | 31 |
Journal | Journal of Applied Logics |
Volume | 10 |
Issue number | 4 |
State | Published - Jul 2023 |
Externally published | Yes |
Bibliographical note
Publisher Copyright:© 2023, College Publications. All rights reserved.
Funding
Supported by FNR Luxembourg INTER-SLANT 13320890, Supported by EU COST Action LITHME CA19102.
Funders | Funder number |
---|---|
European Cooperation in Science and Technology | CA19102 |
Fonds National de la Recherche Luxembourg |