Intelligent Systems Research Centre seminars on IT security

Professor Hassan Kazemian, director of Intelligent Systems Research Centre (ISRC) recently hosted two research seminars on IT security.

Professor Hassan Kazemian, director of Intelligent Systems Research Centre (ISRC) recently hosted two research seminars on IT security.

The first seminar "Individual and group activities simulation in a constrained 3D microworld for capturing behavioural patterns" was presented by a PhD student Pawel Gasiorowski.

It described the methodology of 3D simulation of individual and group behaviour in a closed world - from problem formulation through the development of human behavioural representation model to behavioural pattern analysis. Pavel discussed the process of design and implementation of a 3D simulator, which utilizes concepts, theories and existing solutions known from visual analytics and computer games for capturing relevant data for behaviour pattern analysis. Pavel’s research can be applied to security (e.g. detecting suspicious behaviour), as well as, market research, evacuation planning and many more areas.

The second seminar was presented by Dr Shafi Ahmed, and titled "Using machine learning to detect malicious webpages".

Modern web browsers such as Chrome, Firefox, Internet Explorer and others detect malicious webpages by checking whether they are on 'black lists' that have been created by trusted organisations and volunteers. However, the lists are ineffective because of the frequent-changing nature of webpage.

The research provides an alternative approach to these black lists and shows how to use various machine learning algorithms, both supervised and unsupervised to categorise webpages based on their features such as content, URL information, URL links and screenshots of webpages. The features can be fed into the machine learning models to detect whether a webpage is malicious or not. Prototype tools have been developed to compare and analyse the efficiency of these machine learning techniques with accuracies of up to 98%.

Both seminars were attended by several staff, PhD students and some MSc students, with just over 20 colleagues on each occasion. The seminars’ atmosphere was lively and the exchange of ideas throughout both events was interesting, stimulating and academically challenging.