Back to search

University of Patras

Χρήση τεχνικών τεχνητής νοημοσύνης σε κοινωνικοοικονομικά δεδομένα: εφαρμογή στην Αρχή Προστασίας Ανταγωνισμού (Επιτροπή Ανταγωνισμού)

Abstract

dc:description

The use of Artificial Intelligence (AI) techniques in decision-making and predictive processes constitutes one of the most dynamically evolving research fields in recent years. However, the limited availability of data, as well as the deficiencies or uncertainty that often characterize them, negatively affect the efficiency and reliability of the developed models in this field. Furthermore, the lack of transparency regarding the way results are produced by Machine Learning (ML) methods based on Neural Networks — methods that characterize the connectionist approach —often operate as “black boxes,” restricting interpretability and the reasoning to justify their use. This contradicts symbolic approaches, such as Rule-Based or Case-Based Reasoning methods. The need to understand and interpret the results of these models has led to the development of the field of Explainable Artificial Intelligence (XAI), which aims to create transparent and comprehensible systems that enable users to understand the reasoning behind AI-driven decisions or predictions. To address the above challenges, the combination of different methodologies (hybridization) has been shown to lead to more reliable and effective solutions. In this dissertation (PhD thesis), a hybrid reasoning methodology is analyzed and proposed, combining different methods of the symbolic approach – specifically, Rule-Based Reasoning (RBR) and Case-Based Reasoning (CBR) with Process Mining (PM). This methodology was applied to data from three different domains in order to: i. Develop a Decision Support System for providing personalized care services to individuals with Autism Spectrum Disorders (ASD). The healthcare domain (health data) represents a prominent area for the application of hybrid reasoning systems. With the primary objective of optimizing healthcare management systems, such hybrid models contribute significantly to improving patients’ quality of life and the quality of care provided. ii. Develop the IDS4CLIMA Decision Support System, addressing climate change challenges. The tool integrates the proposed hybrid reasoning (CBR, RBR, and PM) and aims to enhance the capacity for evidence-based and effective decision-making under conditions, where valid scientific knowledge coexists with uncertainty and misinformation. The IDS4CLIMA may also be integrated with other methodologies for assessing environmental costs in monetary terms. iii. Develop the COMP-DST Decision Support System, focused on Competition Law and Policy. Through an integrated, three-level approach, the proposed tool (COMP-DST) systematically structures and utilizes knowledge related to Competition Law and Policy, aiming to assist experts in the analysis and investigation of cases and to improve their efficiency and reliability. Overall, the results demonstrate that hybrid approaches can serve as a critical tool for enhancing the transparency, interpretability, and reliability of decision-support systems in complex and socially sensitive domains.

Degree

thesis:*
Grantor dc:publisher
University of Patras
Year dc:date
2026

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Kaoura, Georgia
  • Καούρα, Γεωργία

Subjects

dc:subject × 29

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/60878

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

Kaoura, Georgia; Καούρα, Γεωργία. Χρήση τεχνικών τεχνητής νοημοσύνης σε κοινωνικοοικονομικά δεδομένα: εφαρμογή στην Αρχή Προστασίας Ανταγωνισμού (Επιτροπή Ανταγωνισμού). University of Patras, 2026. http://hdl.handle.net/10442/hedi/60878