Title: “Statistical methods in cybersecurity”
Date: Monday 29 September 2025, 4.00PM
Location: Online
Speakers: Daniyar Ghani (Imperial College London), Dr Lekha Patel (Sandia National Laboratories), Dr Leigh Shlomovich (Alpha Level Security)
Organizer: Dr Francesco Sanna Passino (Imperial College London), on behalf of the Business Industrial Statistics Section (BIS) and Emerging Applications Section (EAS).
On 29th September 2025, Dr Francesco Sanna Passino hosted an RSS webinar on “Statistical methods in cybersecurity”, jointly supported by the BIS and EAS sections of the Royal Statistical Society. The three speakers (Daniyar Ghani, Dr Lekha Patel and Dr Leigh Shlomovich) are respectively from academia, a US government laboratory, and a small cybersecurity startup. Daniyar Ghani (Imperial College London) has presented statistical models for analysing command logs from honeypots using topic modelling and self-exciting point processes. Lekha Patel (Principal Scientific Machine Learning Research Scientist, Sandia National Laboratories) has discussed real-time anomaly and disturbance detection enabled by cyber-physical digital twins and other threats to cyberphysical systems. Leigh Shlomovich (Head of Research, Alpha Level Security) has focused on statistical methods for classifying alerts, particularly in unsupervised settings, and on developing meaningful, explainable approaches to alert prioritisation.
The webinar was very relevant for RSS members: cybersecurity presents a range of statistical challenges arising from the scale, complexity, and dynamic nature of modern digital systems. Data are often high-volume, heterogeneous, and arrive in real time, requiring methods that are both computationally e[icient and robust to noise. Many cybersecurity applications involve limited or no labelled data, making unsupervised and semi-supervised methods essential. Analysts must often work with evolving threats and incomplete information, requiring models that adapt over time and provide interpretable outputs to support decision-making. These challenges place statistics at the core of developing e[ective, scalable, and explainable solutions for securing digital infrastructure. This webinar has explored some of these challenges, demonstrating how
statistical methods can address key problems in cybersecurity.
The total number of registered attendees as of 29th September 2025, 10am BST was 93. Overall, around 70 attendees joined for part or all of the Microsoft Teams call associated with the webinar.