Back to search

Panteion University of Social and Political Sciences

Εξόρυξη μεγάλων δεδομένων (Data Mining) στην πολιτική επιστήμη: σκιαγραφώντας το πολιτικό πεδίο της Ελλάδας κατά τη διάρκεια της οικονομικής κρίσης

Abstract

dc:description

This dissertation is studying political discourse and focuses on the analysis of electoral behavior. It applies a theoretical approach to the political Left from the perspective of the Lacanian tradition, in combination with computational data mining and analysis of data collected from Twitter, which has been one of the most powerful tools in political communication. The period under investigation covers the three electoral contests in Greece in the years 2015 (January and September) and 2019.For data processing and analysis, we supplemented a specialized dictionary, adapting it to Greek language and ensuring it covered political discourse. Additional-ly, the finding about each election have been visualized, thereby critically complementing the text. In this sense, the visualization does not constitute mere illustration but is an essential part of the research and presentation without which under-standing of the findings and tables remains incomplete. The sentiment analysis conducted according to Plutchik's theory highlighted variations in the intensity of negative emotions. Intensity dynamics may influence electoral behavior and prepare for a shift in roles between government and opposition. Anger and Disgust, along with Disappointment and Shame, dominated all three electoral contests and seem to explain the significant rise of SYRIZA’s appeal in January 2015 when it took power for the first time. In contrast, Contempt appears to be associated with the restoration of the old party system in July 2019.Our research focused on the largest communities within the networks formed each time around the collected tweets. In these, the central figures come from the major parties that occupy the top positions in voters’ choices and participate in the exercise of power. In future research, we intend to focus on medium-sized communities where it is likely to encounter lesser-known public figures from smaller parties and where more intense discussions and debates are likely to take place. The originality of this dissertation lies in the application of theoretical approaches in combination with the adoption of computational methods for the comparative analysis of findings from three successive electoral contests and in the experimentation with new innovative ways of highlighting and visualizing the findings of sentiment analysis.

Degree

thesis:*
Grantor dc:publisher
Panteion University of Social and Political Sciences
Year dc:date
2024

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Messini, Sofia
  • Μεσσήνη, Σοφία

Subjects

dc:subject × 26

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/57137

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Messini, Sofia; Μεσσήνη, Σοφία. Εξόρυξη μεγάλων δεδομένων (Data Mining) στην πολιτική επιστήμη: σκιαγραφώντας το πολιτικό πεδίο της Ελλάδας κατά τη διάρκεια της οικονομικής κρίσης. Panteion University of Social and Political Sciences, 2024. http://hdl.handle.net/10442/hedi/57137