511767 - DATA SCIENCE AND CYBERSECURITY

This course provides a comprehensive introduction to the foundations of data science and its role in the cybersecurity domain. It starts with an overview of the core data science principles and techniques, with particular emphasis on machine learning and deep learning approaches. Moving from these premises, the course examines the role of data science in critical cybersecurity domains, focusing on applications such as Cyber Threat Intelligence (CTI) and anomaly detection. To dive deeper into data science capabilities, the course introduces advanced data-driven methods, including deep learning–based models, Large Language Models (LLMs), and Explainable AI (XAI). These methods are presented in parallel with a critical analysis of their security properties, opportunities and vulnerabilities, ultimately guiding students toward the broader challenges of ensuring the security and trustworthiness of artificial intelligence (AI). At the end of this course the students will learn the main techniques for data science using machine learning and deep learning solutions. They will learn how to leverage data science solutions for cybersecurity applications such as CTI. They will learn techniques to assess the vulnerability of machine-learning algorithms and basic protection strategies. Moreover, they will explore the most common threats to AI-based solutions (such as, data poisoning, model stealing, backdoor attacks, and so forth). Finally, with a reference to the modern international regulations on trustworthy AI, the students will understand the main security aspects to consider in the design of an AI-based solution.