Cybercampus Sweden

Within the framework of Cybercampus Sweden, a PhD project is being carried out named "An Innovative Semi-Supervised Explainable Expert System to Predict Cybersecurity Threats".

Background
The project is based on the fact that cybersecurity environments are characterised by large volumes of unlabeled data, while access to labeled threat data is often limited. Although machine learning has great potential in cybersecurity, traditional methods are highly dependent on extensive, manually labeled datasets. In practice, this is difficult to achieve.

Focus
The project addresses this challenge by using semi-supervised learning, where both labeled and unlabeled data are leveraged to improve threat prediction with minimal human intervention.
A central component of the work is the combination of machine learning and expert knowledge within an explainable framework. By integrating explainable artificial intelligence (XAI), the project develops models that are not only accurate, but also transparent and understandable to human decision-makers. This is crucial for trust, accountability, and practical use in, for example, security organisations and critical infrastructure.

Objectives
The project is expected to contribute to better and earlier detection of cybersecurity threats, thereby strengthening both operational cybersecurity and decision support in Sweden.

Partner
RISE

Contact
Karl Andersson
Professor and Head of Subject, Dean
+46 910 585364
karl.andersson@ltu.se