Data Science for Cyber Security

Dr Pete Burnap from the School of Computer Science and Informatics will outline the School's perspective on cyber security analytics and how various disciplines can contribute.

Location: Room C/2.07, Queen's Buildings, School of Computer Science and Informatics
Date: 9 May 2018 at 15:00
Chair: This seminar will be chaired by Federico Cerutti.

Abstract

2017 saw the launch of the UK Industrial Strategy including Artificial Intelligence (AI). Data Science – the effective blend of data, algorithms, analysis, and real-world problems – will be at the core of this strategy. Cyber security is also high priority for the UK government, with £1.9bn invested in securing UK infrastructure, including the National Centre for Cyber Security who have a focus on active cyber defense.

The harmonisation of the Data Science aspects of AI with cyber security is therefore a major research challenge – one that has been adopted by Cardiff University’s Airbus Centre of Excellence in Cyber Security Analytics. Launched in 2017, the Centre aims to position the UK as a leader in AI for Cyber Security, drawing on academic expertise from computer science, data science, criminology, sociology, law and politics and psychology.

In this talk I will outline our perspective on cyber security analytics and how the various disciplines can contribute. I will then present some of our research on machine learning to detect and predict cyber attacks via malicious executable files – including visualisation through feature maps and sequential learning of ‘cyber kill chains’ through recurrent neural networks.

Finally I will briefly introduce a transformational research project that began in early 2018 – Chatty Factories, funded by the Engineering and Physical Sciences Research Council (EPSRC), aims to merge Internet of Things (IoT) technology with Operational Technology (OT) on the factory floor to create more dynamic and data-driven factories of the future while retaining security and privacy.

Biography