VDVesna Dimitrova

Vesna Dimitrova

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vesna Dimitrova has a total of 39 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Zero-Shot Social Media Crisis Classification: A Training-Free Multimodal Approach

    by Franziska Schwarz, Klaus Dieter Schwarz, Daniel Arias Aranda, Kendrick Bollens, Navaneeth Shivananjappa, Reiner Creutzburg, Vesna Dimitrova

    Rapid classification of social media content during humanitarian crises is essential for effective disaster relief; however, traditional approaches require extensive annotated training data, which are often unavailable during new disasters. This paper presents a training-free, multimodal classification framework that leverages zero-shot vision-language models to analyze disaster-related social media content without task-specific training. The framework employs a two-stage prompt-engineered pipeline using the locally deployable Mistral-Small-3.1-24B-Instruct model, performing binary informativeness detection followed by multiclass categorization into eight humanitarian categories through structured JSON output generation. Evaluation on the CrisisMMD dataset of 18,082 multimodal samples from seven natural disasters demonstrated binary F1 scores above 0.84 for both text and image informativeness detection and weighted F1 scores of 0.61 (text) and 0.72 (image) for humanitarian categorization. The framework generalizes consistently across all disaster types with minimal performance variance (standard deviation below 0.031) and operates entirely on local infrastructure without cloud dependencies, requiring only moderate GPU resources. By eliminating training data requirements, this approach enables immediate deployment during new disasters, demonstrating that zero-shot multimodal classification achieves practically relevant performance for real-time crisis response.


  • Analyzing the Structure of Groupoids of Order 3, 4, and 5 Using PCA

    by Elissa Mollakuqe, Hasan Dag, Vesa Mollakuqe, Vesna Dimitrova

  • Gender and Intersectional Bias in Cybersecurity for Next Generation Wireless Networks

    by Sonay Caner-Yildırım, Isabella Corradini, Vesna Dimitrova, Valeria Loscri, Miranda Harizaj

    The advancement of wireless networks towards sixth-generation (6G) and pervasive artificial intelligence (AI) underscores critical barriers to trustworthy cybersecurity, notably algorithmic bias, workforce underrepresentation, and the neglect of intersectional gender perspectives. Current cybersecurity research predominantly emphasizes algorithmic performance, often overlooking broader societal impacts such as equity and inclusivity. This paper addresses this gap by synthesizing interdisciplinary lessons from emerging technology deployments, critically examining risks inherent in AI-driven 6G security, and articulating the urgent need for an “Equity by Design” approach. We introduce a comprehensive roadmap to embed intersectional equity into cybersecurity processes—from policy-making and technical development to real-world deployment. Utilizing the BEiNG-WISE COST Action as a best-practice case study, we demonstrate how explicit gender-inclusive and participatory methodologies can effectively operationalize equity, transparency, and diversity. Our findings illustrate that integrating social and technical dimensions is essential for building secure, resilient, and equitable digital futures.

  • Open Access

    Advancing Image Spam Detection: Evaluating Machine Learning Models Through Comparative Analysis

    by Mahnoor Jamil, Hristina Mihajloska Trpcheska, Aleksandra Popovska-Mitrovikj, Vesna Dimitrova, Reiner Creutzburg

    Image-based spam poses a significant challenge for traditional text-based filters, as malicious content is often embedded within images to bypass keyword detection techniques. This study investigates and compares the performance of six machine learning models—ResNet50, XGBoost, Logistic Regression, LightGBM, Support Vector Machine (SVM), and VGG16—using a curated dataset containing 678 legitimate (ham) and 520 spam images. The novelty of this research lies in its comprehensive side-by-side evaluation of diverse models on the same dataset, using standardized dataset preprocessing, balanced data splits, and validation techniques. Model performance was assessed using evaluation metrics such as accuracy, receiver operating characteristic (ROC) curve, precision, recall, and area under the curve (AUC). The results indicate that ResNet50 achieved the highest classification performance, followed closely by XGBoost and Logistic Regression. This work provides practical insights into the strengths and limitations of traditional, ensemble-based, and deep learning models for image-based spam detection. The findings can support the development of more effective and generalizable spam filtering solutions in multimedia-rich communication platforms.


  • INTERNATIONAL CYBERSECURITY COOPERATION: THE CYBERMACS ERASMUS MUNDUS INITIATIVE FOR A SECURE GLOBAL DIGITAL FUTURE

    by Reiner Creutzburg, Vesna Dimitrova, Ivan Chorbev, Ebru Dilan, Ayhan Gücüyener, Hasan Dağ

  • Open Access

    Mathematical Foundations and Implementation of CONIKS Key Transparency

    by Elissa Mollakuqe, Hasan Dag, Vesna Dimitrova

    This research paper explores the CONIKS key management system’s security and efficiency, a system designed to ensure transparency and privacy in cryptographic operations. We conducted a comprehensive analysis of the underlying mathematical principles, focusing on cryptographic hash functions and digital signature schemes, and their implementation in the CONIKS model. Through the use of Merkle trees, we verified the integrity of the system, while zero-knowledge proofs were utilized to ensure the confidentiality of key bindings. We conducted experimental evaluations to measure the performance of cryptographic operations like key generation, signing, and verification with varying key sizes and compared the results against theoretical expectations. Our findings demonstrate that the system performs as predicted by cryptographic theory, with only minor deviations in computational time complexities. The analysis also reveals significant trade-offs between security and efficiency, particularly when larger key sizes are used. These results confirm that the CONIKS system offers a robust framework for secure and efficient key management, highlighting its potential for real-world applications in secure communication systems.

  • Application of Machine Learning in Intrusion Detection Systems

    by Milena Gjorgjievska Perusheska, Vesna Dimitrova


  • Applications of Quasigroups in Cryptography and Coding Theory

    by Smile Markovski, Verica Bakeva, Vesna Dimitrova, Aleksandra Mileva, Aleksandra Popovska-Mitrovikj, Hristina Mihajloska Trpcheska


  • Open Access

    Classification and Analysis of Cyberattacks in the Western Balkans Countries - with a focus on the treatment of policy violations and strategies that states use to protect against attacks

    by Elissa Mollakuqe, Vesna Dimitrova

    Rapid technological developments have also affected the countries of the Balkans, where as a result many hackers and other attackers have attempted to attack computer systems and data in these countries as well. The initial attacks that directed you to these places were "phishing" attacks where there was an attempt to suspend the work of online systems, especially in Public Institutions. But over time, especially in the last 3 years, the biggest attacks belong to DDoS attacks, which have caused many problematic situations and an instability of the systems in these countries. Cyber ​​attacks are normal in the era we live in now, but the reasons for these attacks still vary from the most diverse, such as: political blackmail, blackmail of other kinds or even blackmail for monetary benefits. Our research considers 7 countries of the Balkans: Kosovo, North Macedonia, Albania, Serbia, Montenegro, Bosnia and Herzergovina, and Croatia. The main analysis focuses on the treatment of policies to protect against cyber attacks and strategies that these countries have with the aim of protecting against cyber attacks, which were on a large scale during the last period. All data has been collected during real time and the countries are ranked according to the level of exposure to the risk. The purpose of this research is to know the current situation in the Balkans and to identify the types of attacks directed at these countries, also some investigations into the current strategies have been proposed. The paper uses methods of analysis and synthesis, comparative methods, as well.