National Technical University of Athens (NTUA)
A social and emotional learning methodology leveraging social network analysis, knowledge graphs and generative artificial intelligence
Abstract
dc:descriptionQuite a lot of effort has been put into developing Social and Emotional Learning (SEL) programs in K-12 schools (from kindergarten to high school). However, there are many challenges that hamper the wide adoption of SEL methodologies, starting from the existence of a plethora of heterogeneous frameworks that apply SEL in many different ways, the lack of open Emotional Intelligence (EI) models, the high workload among teachers, and the difficulty of implementing holistic methodologies that simultaneously integrate assessment, planning, implementation, and monitoring. Information and Communication Technologies (ICT) can help teachers with this task by automating routine tasks such as secure data collection, continuous progress tracking, planning of SEL activities, and real-time feedback, freeing up their time and energy for more meaningful student participation and personalized learning during the implementation of the SEL program. By simplifying workflows, ICT not only reduces administrative burden, but also empowers teachers with focused knowledge in the domain of social and emotional learning, enhancing at the same time their technological skills. The objective of this dissertation is to propose a comprehensive ICT-enabled SEL methodology that covers the entire lifecycle of a SEL program in primary and secondary schools. Unlike existing approaches that focus on specific phases, the proposed methodology spans from teacher training and student assessment to the design of targeted interventions, the formation of working subgroups, and the analysis of program impact, providing for each phase both a theoretical foundation and practical open-source digital tools to support educators and students. Specifically, it combines concepts from the field of Emotional Intelligence and sociometry for student assessment, Social Network Analysis techniques for the visualization and interpretation of classroom social dynamics, multi-criteria partitioning algorithms for the formation of working subgroups that take into account students' emotional profiles and social relationships, and Reinforcement Learning techniques for the recommendation of SEL activities. Furthermore, a Knowledge Graph is developed to store and interconnect anonymized data from real classrooms under a common semantic framework. This includes student assessment results, records of the SEL activities implemented, and data on their impact on classroom climate and socio-emotional competences. This Knowledge Graph serves as the foundation for an agentic digital assistant that leverages Generative AI and Graph Retrieval-Augmented Generation (GraphRAG) techniques to provide educators with evidence-based answers to theoretical SEL questions, personalized interpretation of classroom assessment results, and targeted activity recommendations and learning pathways tailored to the identified needs of students at both individual and collective level. The main research challenges this dissertation aims to address include the need for safe and ethically responsible management of sensitive personal data collected during student assessment, the absence of open and semantically enriched Emotional Intelligence models covering all age groups from early childhood to adulthood, and the need to integrate state-of-the-art technologies in a way that ensures usability for educators without specialized technical knowledge. Finally, the dissertation aims to empirically demonstrate the effectiveness of the proposed methodology through evidence-based outcomes in a wide set of primary and secondary schools, acknowledging that its broad adoption depends largely on the trust that educators develop toward the tools and the methodology. Based on the above, it can be concluded that effective design of SEL programs requires knowledge and collaboration across multiple scientific disciplines. The present dissertation is built on the intersection of three distinct research domains, namely Emotional Intelligence, Education Sciences, and Information and Communication Technologies. The core contribution of this dissertation lies in the integration of these domains into a coherent and pedagogically grounded methodology that is both theoretically sound and practically applicable in real school settings.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Fotopoulou, Eleni
- Φωτοπούλου, Ελένη
Subjects
dc:subject × 50- Μεθοδολογία Κοινωνικής και Συναισθηματικής Μάθησης (ΚΣΜ)
- Εκπαίδευση Υποστηριζόμενη από ΤΠΕ
- Ανάλυση κοινωνικών δικτύων
- Γράφος Γνώσης
- Αξιολόγηση Συναισθηματικής Νοημοσύνης
- Συλλογική Συναισθηματική Νοημοσύνη
- Ψυχομετρικό εργαλείο
- Κοινωνιομετρία
- Σημασιολογικός εμπλουτισμός
- Οντολογία
- Σύστημα Σύστασης με Ενισχυτική Μάθηση
- Πολυκριτηριακή Κατάτμηση Κοινωνικών Ομάδων
- Παραγωγική τεχνητή νοημοσύνη
- Τεχνητή Νοημοσύνη Πολλαπλών Πρακτόρων
- Επαυξημένη ανάκτηση πληροφορίας βασισμένη σε Γράφους Γνώσης
- Εργαλεία Ανοικτού Κώδικα για την Εκπαίδευση
- Αποθετήριο Δραστηριοτήτων ΚΣΜ
- Ηθική Τεχνητή Νοημοσύνη στην ΚΣΜ
- Social and Emotional Learning (SEL) Methodology
- ICT-enabled Education
- Social Network Analysis (SNA)
- Knowledge Graph
- Emotional Intelligence Assessment
- Collective Emotional Intelligence
- Psychometric Tool
- Sociometry
- Semantic Alignment
- Ontology
- Reinforcement Learning driven Recommendation System
- Multi-criteria Social Group Partitioning
- Generative AI
- Agentic AI
- Multi-agent systems
- Graph Retrieval Augmented Generation (GraphRAG)
- Open-Source Educational Tools
- K-12 Education
- SEL Activities Repository
- Ethical AI in SEL
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Εκπαίδευση
- Κοινωνικές Επιστήμες
- Εκπαίδευση και Έρευνα της εκπαίδευσης
- Electrical Engineering, Electronic Engineering, Information Engineering
- Engineering and Technology
- Communication engineering and systems, Telecommunications
- Education
- Social Sciences
- Education and Educational Research
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/62181