A special edition of iMEdD for the general elections in Greece
A collaboration between humans and artificial intelligence to analyze the pre-election campaign speeches of Greece’s main political leaders: what issues they raise in the public discourse, what sentiments they convey, and to what extent polarization and populism can be detected in their rhetoric.
Sentiment analysis per party leader and speech
Note: The analysis refers to the sentiment detectable in the political speeches themselves and is not at all related to audience emotions.
* The speeches of Kyriakos Velopoulos are statements in press conferences he has given. Therefore, these analyses cannot be taken as directly comparable to those of other party leaders that are given in the public space.
Sentiment analysis during the election period for two different party leaders
Sentiment analysis during each speech between two party leaders
Sentiment analysis by topic in the speeches of party leaders
Note: If a particular topic is not mentioned in a party leader’s speech, the corresponding space in our analysis will be left blank, and therefore, there is no sentiment analysis available for visualization.
How did political leaders address the issue of migration and refugees prior to the tragic shipwreck off the coast of Pylos?
A slight rise in polarizing rhetoric and migration crisis mentions in political leaders’ discourse after the tragedy off the coast of Pylos
The first inklings of the political leaders’ rhetorical strategy during the new campaign period: what they are keeping from the previous election period and what is changing
See the analyses of all the speeches of the leaders of the parliamentary parties before the elections of 21 May 2023.
What percentage of each speech is devoted to the presentation of the programmatic agenda of each party and what is the extent of the criticism of the opponents?
The topics each party leader addresses during their speech
By systematically identifying and categorizing the primary topics discussed by political leaders, the project is able to accurately assess their priorities, policy proposals, and overall campaign messaging.
The pre-election discourse, the post-election environment and artificial intelligence in political analysis
Nine experts from the fields of political science and communication, computer science and journalism write about the post-election political scene, the campaigns leading up to the elections and the contribution of artificial intelligence to political analysis and journalism.
The working methodology for analyzing campaign speeches in the project -from data collection to data analysis.
Polarization and populism during the pre-election period
The following linear step diagram illustrates the evolution of polarization or populism during the election period, as represented by each political leader. On the vertical axis, the percentage of each campaign speech containing elements of moderate or high polarization or populism from each political leader’s discourse is displayed. The horizontal axis presents the corresponding speeches, from the official declaration of elections in Greece to the last available election speech of each party leader.
Polarization and populism per campaign speech
In the step chart, the evolution of the level of polarization or populism during each speech is depicted, with the number of paragraphs shown on the horizontal axis. Each paragraph, depending on the value of the polarization and populism indicators it has received, is placed in a category:
- no or low level (polarization or populism respectively) if it has a value from 0 to 0.5,
- medium level if it has a value from 0.51 to 0.8,
- high level if it has a value from 0.81 to 1.
Populism is not equivalent to politicians’ falsehoods, promises, or deceitfulness. It is not synonymous with demagoguery, propaganda, or manipulation.
Differentiating competition, division and polarisation in political discourse.
The political discourse analysis is highly useful for understanding the dynamics of political competition.
How to read the radial dendrogram below
The radial dendrogram visually represents the entities (names, organizations, geographical areas, etc.) extracted from the speeches of the six political leaders under study (Named Entity Recognition, NER). At the centre of each speech, there is a node which branches into sub-nodes corresponding to the different groups of entities named in the election speeches.
To “read” this visualization, start from the central node of a speech and follow the branches to explore the named entities and their connections. The numbers in parentheses indicate the frequency of mention for each entity in the speech: respectively, the larger the entity node, the more times the term in question has been mentioned.
Who is referring to whom
We count the number of references each leader makes to his opponents
How many “will” has every political leader said in this campaign?
We count the number of “will” heard from each political leader per 10,000 words spoken
*The speeches of Yanis Varoufakis are not analysed during the second pre-election period, since the project covers the speeches of the political leaders representing the last parliamentary parties each time.
The campaign trails of the political leaders
Who worked on this project
Idea & Project Coordination: Thanasis Troboukis, Kelly Kiki (iMEdD)
Journalistic Research/Analysis: Nota Vafea, Katerina Voutsina, Stefania Ibrishimova, Athina Thanasi, Kelly Kiki, Chrysoula Marinou, Thanasis Troboukis, Georgios Schinas (iMEdD)
IT Support: Christos Nomikos, Nikos Sarantos (iMEdD)
Scientific Advisor on Political Theory: Antonis Galanopoulos, PhD Candidate at the School of Political Sciences, Aristotle University of Thessaloniki
Software Development/ Data Analysis: Pavlos Sermpezis, Stelios Karamanidis, Dimitrios-Panteleimon Giakatos, Ilias Dimitriadis (Datalab, School of Informatics, Aristotle University)
Datalab Director (School of Informatics, Aristotle University): Professor Athena Vakali
Translation: Anatoli Stavroulopoulou, Tina Katoufa