Svetlana Petrova Stefanova

Dr. Svetlana Petrova Stefanova

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 November 2025 in Sustainability
Verified
Journal Article
Open access
PDF file
Ana TodorovaAna Todorova
Ana Todorova
IKIrina Kostadinova
Irina Kostadinova
SSSvetlana Petrova Stefanova
Svetlana Petrova Stefanova
This article explores the potential of an integrated pedagogical approach that combines project-based learning (PBL) with Internet of Things (IoT) technologies to achieve the United Nations’ Sustainable Development Goals (SDGs). Within the context of Education for Sustainable Development (ESD), this model transforms students from passive consumers of information into active agents of change. The study demonstrates that leveraging IoT sensors enables students to tackle pressing and complex sustainability challenges by engaging them in a comprehensive problem-solving cycle—from collecting real-world data to developing innovative solutions. By analysing the existing scientific literature, the approach is shown to significantly improve critical thinking skills, systems thinking, creativity, and teamwork. The study also proposes a new conceptual framework (hypothesis), the EcoHabits model, whose effectiveness remains to be empirically validated. This model demonstrates IoT’s potential to enhance sustainability competencies, improve health literacy, and promote individual and collective behaviour change. Despite its significant pedagogical benefits, the article identifies key implementation challenges, including the need for adequate teacher training and community collaboration. In conclusion, this innovative framework offers a promising solution for preparing future generations to address global issues and become active, responsible citizens of the 21st century.

11 June 2021
Verified
Conference Paper
Closed access
NNNeyko Neykov
Neyko Neykov
SSSvetlana Petrova Stefanova
Svetlana Petrova Stefanova
Collecting and processing the knowledge in a usable way is a quite important activity for almost every company nowadays. The importance of creating innovative products raises, no matter what the domain is in which the company operates. This cannot be done without building and supporting a knowledge base with existing knowledge and extending it with new harvested or generated knowledge. In this article we provide a review of known types of focused crawlers, which can be used by companies to collect knowledge in the global internet space as well as in the intranet and build their own knowledge base.


5 December 2025 in BRAIN. BROAD RESEARCH IN ARTIFICIAL INTELLIGENCE AND NEUROSCIENCE
Verified
Journal Article
Open access
PDF file
Ana TodorovaAna Todorova
Ana Todorova
IKIrina Kostadinova
Irina Kostadinova
SSSvetlana Petrova Stefanova
Svetlana Petrova Stefanova
In recent years, interest in developing emotional and social skills in artificial intelligence has grown significantly. The present study aims to explore the options for measuring the emotional and social aspects of artificial intelligence, as well as to analyse the generated results and the formulated conclusions and generalisations. An analysis of existing and widely used chatbots with artificial intelligence was conducted, and the emotional and social intelligence of AI was assessed using a non-standardised test questionnaire. The findings provide essential conclusions, not about the possibilities for measuring the intrapersonal and social skills of AI, but rather about the current state of its development and its ability to generate emotional and social insights. The authors also establish a correlation between the affective and interpersonal intelligence demonstrated by AI and collective (global) emotional intelligence as a concept. In conclusion, the article discusses the potential applications of these results in various fields, including enhancing communication capabilities between humans and machines, among humans themselves, and even between machines. This study provides essential data and directions for future research and applications in the field of human emotional and social intelligence, as well as AI, acknowledging the use of free chatbot versions and a non-standardised questionnaire. In this sense, the insights obtained enrich both the theoretical and applied knowledge in the fields of information, communication technologies, and social sciences, including management and human resources.

30 June 2026 in Electronics
Verified
Journal Article
Open access
PDF file
RKRaid Anis Kerkatou
Raid Anis Kerkatou
HBHacene BELHADEF
Hacene BELHADEF
AEAICHA EUTAMENE
AICHA EUTAMENE
SSSvetlana Petrova Stefanova
Svetlana Petrova Stefanova
Network intrusion detection systems (IDS) operate in high-dimensional feature spaces under evolving attack patterns and asymmetric misclassification costs, where false negatives represent a critical security risk. Reinforcement learning (RL) offers a natural mechanism for encoding domain-specific misclassification costs directly into the learning signal through reward shaping, enabling cost-sensitive policy optimization in adaptive streaming environments. However, the integration of variational quantum models into RL-based IDS remains insufficiently explored. This work investigates a variational quantum reinforcement learning (VQRL) framework for intrusion detection, in which parameterized quantum circuits are employed to model the policy function. We adopt an RL formulation primarily as a principled cost-sensitive optimization approach rather than to exploit sequential state dependencies, and we employ Instantaneous Quantum Polynomial (IQP) embedding as a quantum feature encoding strategy. The study analyzes how embedding expressivity interacts with varying levels of dimensionality reduction via principal component analysis (PCA) on the CICIDS2017 dataset. Experiments demonstrate that VQRL-IQP achieves high recall and reduces false negative rates in moderately high-dimensional feature spaces compared to a classical RL baseline. This improvement is accompanied by an increase in false positive rates, reflecting a trade-off shaped jointly by the reward structure and the structural properties of IQP encoding. Statistical validation across five independent runs confirms the consistency of these trends. Importantly, no general quantum advantage in accuracy or computational efficiency is claimed; rather, the results indicate that VQRL-IQP offers a distinct error trade-off that is operationally valuable in security-critical scenarios where minimizing missed attacks is the primary objective.