The goal of our research into cognitive security is to accelerate the move away from the current generation of reactive systems to a proactive generation of cognitive systems. To achieve this, we look at ways of distilling enormous volumes of structured and unstructured data into information, and then into actionable knowledge to enable continuous security and business improvement. This involves the use of automated, data-driven security technologies as well as techniques and processes that enhance cognitive systems with the highest level of context and accuracy.
The term Internet of Things (IoT) has come to denote the wide-spread deployment of sensors and actuators with ubiquitous inter-connectivity, potentially to the Internet, to monitor physical systems and critical infrastructure such as electrical utilities, transportation systems, SCADA systems, factories and buildings.
The security of such systems has unfortunately not attracted as much attention as it deserves. The spate of recent attacks that have been realized on these embedded sensors bears testimony to their ubiquity, but also raises serious concerns about their security. These devices, by virtue of their deployment, can be used to cripple critical infrastructure that was once considered invulnerable.
Our research addresses a number of aspects of IoT security such as the passive data collection and the reconstruction of SCADA protocols in order to allow the cognitive detection of security relevant events.
Social media security analytics
Public social media data is not only a source for public sentiment, it contains insight that can be extracted and used to enrich cyber security defences. Extracting actionable insight is another area of research where Big Data technologies are used to collect social media and public data in order to generate relationships as graphs.
We evaluate ways to augment these graphs and strategies to enrich them with the aim of improving the capability of our analytics.