Back to search

University of the Aegean

Monitoring of social, economic and environmental state of regions and prediction of changes using machine-learning approaches trained with historical data

Abstract

dc:description

By testing and using Machine Learning (ML) and data related approaches and tools, this research attempts to enhance regional science towards sustainable development. Specifically, this Thesis aims to contribute to the study of regions’ resilience, by introducing the use of relevant ML models that can extract useful information from economic, environmental and social indicators and use this data for the development of advanced features such as predicting the performance of regions in disasters (US counties case presented) or correlating strategic aspects of Circular Economy (CE) with regions’ resilience (EU regions case presented). Regarding the ML approaches, this thesis aims:•To create and demonstrate models through Machine Learning that can predict the performance of a region in disasters and can find the parameters/factors (confirm obvious from classic regional approaches or show hidden from data) that affect the resilience of the regions.•To mainly contribute via models’ demonstration, to a methodology, offering useful guidelines to pave the way on how the various regional factors (characteristics and properties of the regions, related to economic, social and environmental dimensions, such as demographics, GDP, the dominant industries, other statistics related to employment, entrepreneurship, land use, commuting, transportation, health, education etc.) can be translated and be processed by data and ML/AI tools and techniques so different models can be developed. It should be clearly noted that this thesis aims to contribute to regional science. Although it uses technical and digital functionalities by applying ML methods, these are treated as tools (even as "black boxes" in some cases) to enhance regional science, and not as areas of research. Therefore, no technical progress is sought in the fields of ML and data science and modeling. Specifically, this Thesis contributes to regional science by:• Creating ML models for the US counties and EU regions (NUTS2) that predict their performance, thus a side of their resilience, under crisis periods.• Exploring and collecting heterogeneous regional data, harmonizing them and transforming them into usable data by ML techniques.• Proposing an overall methodology of ML models for regional study, highlighting the stages of the model creation process.• Proposing and describing the exploitation of the models and the methodology on policy implications and decision support processes.• Designing a decision support tool that can be developed to support regional authorities, policy makers and other stakeholders to analyze and strengthen their regional status and develop strategies, action plans and policies.Also, as an extra and secondary objective, this study aims to deal with, discuss and present how existing or new methodologies, technological trends, approaches -e.g. the major approach of circular economy or digitalization- affect the resilience of the regions and contributes to its measurement, its monitoring and its improvement via valuable conclusions and future research and innovation recommendations.

Degree

thesis:*
Grantor dc:publisher
University of the Aegean
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Manousiadis, Charalampos
  • Μανουσιάδης, Χαράλαμπος

Subjects

dc:subject × 20

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/60484

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

Manousiadis, Charalampos; Μανουσιάδης, Χαράλαμπος. Monitoring of social, economic and environmental state of regions and prediction of changes using machine-learning approaches trained with historical data. University of the Aegean, 2025. http://hdl.handle.net/10442/hedi/60484