Panteion University of Social and Political Sciences
Εφαρμογή προηγμένων μεθόδων στατιστικής και αλγόριθμων εξόρυξης πληροφορίας για την ανάλυση κοινωνιοψυχολογικών ερευνών με σκοπό τη διερεύνηση στάσεων, αντιλήψεων και προτύπων συμπεριφοράς
Abstract
dc:descriptionThis PhD dissertation aims to add an additional methodological approach for the analysis of social research. This approach is based on Data Mining that involves methods at the intersection of artificial intelligence, statistics, and database systems. Data mining aims at revealing underlying structures within data and summarising them with innovative and intelligible ways. Algorithms of Data Mining are widely used in many scientific fields and are usually preferable when big databases are available. However, their use in research of social psychology, where multidimensional concepts are involved, is not common. Due to the infiltration of technology in everyday life the recording of big data is nowadays reachable, while the need of understanding complex social issues remains, hence, the implementation of such algorithms in social psychology is a new methodological approach that can be used. The approach of data mining that is presented in this dissertation does not intend to be compared with the typical methods of statistics. Both approaches can be used complementarily, not only in order to overcome their restrictions but also to broaden the confines of analysis. For this dissertation, we implemented decision tree algorithms for the study of four topics. The first topic concerned the way that mentality and political self-placement mould social tolerance in phenomena of physic and symbolic violence. The second one focused on the way that mentality and political self-placement sets xenophobic perceptions. For the third topic, an approach of the national consciousness through the placement of Greece on the global map is presented, and compares Greece with other countries. The last topic is dedicated to the definition of European Union (EU) by Greek people. Using a set of questions about the present and the future of the EU, algorithms of Data Mining were used in order to reveal the patterns that form how Greek people perceive EU in a period during which Greece is beginning a new route in a globalised society. The application of data mining algorithms and multidimensional statistical methods resulted in patterns that underlie the examined issues and that were used for the classification of social subjects into predefined categories with discrete perceptions.
Degree
thesis:*- Grantor dc:publisher
- Panteion University of Social and Political Sciences
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Massou, Efthalia
- Μάσσου, Ευθαλία
Subjects
dc:subject × 24- Αλγόριθμοι εξόρυξης πληροφορίας
- Δέντρα αποφάσεων
- Αλγόριθμοι ταξινόμησης
- Πρότυπα
- Φυσική και συμβολική βία
- Νοοτροπία
- Ξενοφοβία
- Πολιτική αυτοτοποθέτηση
- Εθνική ταυτότητα
- Ευρωπαϊκή Ένωση
- Data mining
- Decision trees
- Classification algorithms
- Patterns
- Physic and symbolic violence
- Mentality
- Xenophobia
- Political self placement
- National consciousness
- European Union
- Κοινωνικές Επιστήμες
- Ψυχολογία και Γνωσιακή Επιστήμη
- Social Sciences
- Psychology and Cognitive Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/40445