Welcome to the Cyber Analytics lab at Delft University of Technology. We perform research in Artificial Intelligence and Data Science for Cyber Security and Software Engineering. Our main aim is to get knowledge from software data and use that for analysis, anomaly detection, and classification/intrusion detection. That is why our focus is on developing learning algorithms for white-box models that can be understood by developers and security analysts, such as decision trees, temporal heatmaps, and state machines (or deterministic Markov models). Futhermore, we make them learn from streaming data and robust against evasion. Key topics we study are: State Machine Learning, Automated Reverse Engineering, Network Traffic Analytics, Attacker Behaviour Modelling, Declarative/Optimal Machine Learning, Fuzzing/Symbolic Execution, and Adversarial Machine Learning.
Recent News
- Our student Tom Catshoek together with Sicco Verwer won the LTL track of the Rigurous Examination of Reactive Systems (RERS 2020) challenge!
- We have a new twitter handle @analytics_cyber. Follow us to receive updates about our research endeavours!
- Sicco Verwer and Toon Calders have received the Test of Time Award for their paper Three naive Bayes approaches for discrimination-free classification at ECML-PKDD 2020.
- Our paper entitled “The Robust Malware Detection Challenge And Greedy Random Accelerated Multi-Bit Search” by Sicco Verwer, Azqa Nadeem, Christian Hammerschmidt, Laurens Bliek, Abdullah Al-Dujaili, Una-May O’Reilly has been accepted to Artificial Intelligence and Security (AISec) workshop co-located with ACM CCS 2020.
- Our paper entitled “Beyond Labeling Using Clustering To Build Network Behavioral Profiles Of Malware Families” by Azqa Nadeem, Christian Hammerschmidt, Carlos H. Ganan, Sicco Verwer will appear as a chapter in the Malware Analysis using Artificial Intelligence and Deep Learning (MAAIDL) book by Springer.