Διδακτορικές Διατριβές / Doctoral Theses

Μόνιμο URI για αυτήν τη συλλογή

https://apothesis.hmu.gr/handle/123456789/10967

Περιηγούμαι

Πλοήγηση Διδακτορικές Διατριβές / Doctoral Theses ανά Συγγραφέας "Papadakis, Nikolaos"

Τώρα δείχνει 1 - 2 of 2
  • Μικρογραφία εικόνας
    Τεκμήριο
    Smart-home security using the Internet of Things threats and countermeasures
    (ΕΛΜΕΠΑ, Σχολή Μηχανικών (ΣΜΗΧ), Τμήμα Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών, 2025-10-16) Vardakis, Georgios; Βαρδάκης, Γεώργιος; Papadakis, Nikolaos; Παπαδάκης, Νικόλαος
    With the rapid advancements in the Internet of Things (IoT) landscape, the integration of interconnected devices within residential environments has catalyzed the widespread emergence of smart homes. While these developments facilitate enhanced interoperability and automation, they simultaneously introduce intricate security challenges stemming from device heterogeneity and the complexity of underlying communication protocols. This dissertation systematically maps the cybersecurity landscape of smart home ecosystems, advancing beyond generalized perspectives to provide an in-depth examination of critical IoT device categories including digital voice assistants, smart televisions, augmented and virtual reality systems (AR/VR), intelligent locks, and environmental sensors.The study delineates representative threat vectors such as unauthorized system access, side-channel attacks, personal data exfiltration, firmware tampering, and device compromises. Furthermore, it critically evaluates current security mechanisms and methodologies designed to mitigate these risks, encompassing lightweight cryptographic protocols, multi-factor authentication (MFA) schemes, intrusion detection systems (IDS), and anomaly detection techniques, with the objective of fortifying system resilience against sophisticated adversarial behaviors.Beyond the purely technical facets, this work underscores the pivotal role of user awareness and education in sustaining secure smart home environments. Despite technological improvements in construction standards and living conditions, subjective perceptions of safety remain frequently insufficient, prompting intensified research efforts towards advanced protective technologies. Heightened insecurity, particularly in densely populated urban centers, has stimulated the application of Artificial Intelligence (AI) and Machine Learning (ML) methods for the development of intelligent threat prevention and response systems, thereby enhancing both physical and cyber security perceptions. Within this context, applications leveraging Deep Learning (DL) methodologies have emerged, transforming conventional residences into automated, adaptive, and secure infrastructures. The dissertation highlights DL’s dual contributions: enhancing cybersecurity in smart homes and enabling interdisciplinary breakthroughs in domains such as medicine, where remarkable outcomes are observed. Detailed exposition is provided on the backpropagation algorithm employed in both linear and nonlinear artificial neural networks, including a case study of the XOR problem as a classical demonstration of nonlinearity.Machine Learning demonstrates high efficacy in human activity analysis, particularly in facial recognition, which constitutes a foundational component for personalized management and security in smart home settings. The imperative for real-time processing of voluminous data, coupled with substantial computational demands, necessitates the deployment of fog and cloud computing paradigms to efficiently support image processing, face identification, and human silhouette recognition. Collectively, these subsystems form a comprehensive security framework grounded in DL architectures and ML-trained models. Based on the findings of this research, it is substantiated that the application of such technologies significantly advances the realization of fully secure residences, addressing an urgent societal need especially in light of escalating crime rates. Finally, the dissertation proposes avenues for future research aimed at further reinforcing IoT-enabled smart home security. Recommendations include the development of robust unified security standards, enhanced device authentication protocols, and the incorporation of intelligent intrusion detection mechanisms employing anomaly analytics and continuous learning. Addressing these technical challenges is essential for actualizing the vision of fully operational, secure, and threat-resilient IoT-enabled smart homes that safeguard user privacy and digital sovereignty.
  • Μικρογραφία εικόνας
    Τεκμήριο
    Workload based summaries for knowledge graphs
    (ΕΛΜΕΠΑ, Σχολή Μηχανικών (ΣΜΗΧ), Τμήμα Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών, 2024-10-16) Vasileiou, Ioannis; Βασιλείου, Ιωάννης; Papadakis, Nikolaos; Παπαδάκης, Νικόλαος
    This dissertation delves into the advancements in semantic summarization and user centric exploration of Knowledge Graphs (KGs), driven by the rapid expansion of interconnected data. Semantic summaries have become critical tools for distilling vast datasets into manageable sizes, optimizing query answering, indexing, and visualization. Recent developments in structural semantic summaries have focused on extracting central nodes from the semantic graph, exploiting several graph centrality measures, then linking them and presenting them as summaries. Those summaries can then be used among others to optimize query answering as the size of the graph is drastically reduced. However, as the semantic graphs are heterogeneous, using variations of centrality measures for selecting parts of the graph to be used as a summary, generates summaries with limited benefits for query answering. Leveraging user query workloads has the potential to offer tangible benefits to this direction, as they can offer unique insights on trends and user interests as they evolve over time. To this direction, this dissertation starts exploring workload-based summaries by selecting nodes based on their frequency in query workloads. This drastically improves the usefulness of the result summaries in terms of query coverage. Then it explores how utilizing query logs and Large Language Models (LLMs) can lead to the automatically generation of FAQs, enabling users to rapidly understand the contents of an entire KGjust visiting a set of questions and their answers in textual format. In parallel, we explore shifts in user interests over time using query logs and language models, facilitating users' visualization and understanding of these evolving interests. Then we focus on how to construct personalized summaries that adapt to individual user preferences. We again exploit query logs selecting queries similar to the interests of the user and generate summaries maximizing coverage for user queries, dominating all baselines and competitors. Finally we focus on how workload-based summaries can be used for the generation of compact structures that can be used as a caching mechanism to rapidly provide a first answer to user queries before answering their queries in full. We demonstrate that such summaries are both practical, as they can be trivially constructed and retained in main memory, and also of high benefit as they can significantly optimize the time required for the first results of user queries. By integrating these innovative approaches, this dissertation aims to advance the field of semantic summarization and user-centric KG exploration, fostering more effective and efficient data exploration in increasingly interconnected environments.