Gamification and AI guidance for training in Cyber ranges

Details

Supervisors

Legay, Axel

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] en sciences informatiques, à finalité spécialisée

Abstract

Cybersecurity training is essential for everyone who interacts with digital systems, as human error is a primary vector for cyber attacks. With the rapid emergence of new vulnerabilities due to technological advancements, continuous and role- specific training is crucial. However, maintaining motivation for ongoing education, especially among non-specialists, remains challenging. Research highlights that gamification can significantly enhance engagement and motivation, while a lack of adaptive learning difficulty can lead to frustration and disengagement. This thesis presents a comprehensive cybersecurity training platform designed to address these challenges. The platform integrates multiple gamification elements to increase engagement and includes two key guidance techniques to support users with different levels of expertise. First, an adapted recommendation system tailored to the specific nature of cybersecurity training scenarios provides personalized guid- ance outside of scenarios. Second, a chatbot powered by advanced large language models (LLMs) offers real-time support within scenarios, helping users navigate tasks and answering their questions. The platform currently includes two functional scenarios focused on web vulnerabilities. While these scenarios are specific, the use of cyber ranges as a core technology enables the creation of a wide variety of realistic and diverse training environments. By covering four primary research areas: cyber ranges, gamification, recom- mender systems, and LLMs, this work does not aim to push the boundaries of any single field. Instead, it innovates by combining these domains in a way that sustains motivation and provides the necessary guidance for users at different levels of expertise.