Research at The Institute

The institute applies their research to real-world issues whilst bridging the gap between industry, government and academia. Fundamental research areas currently include data-centric engineering, high-performance computing, cyber-security, smart cities, health, the economy and data ethics.

The research was never this prolific though and the beginnings of the centre were tough. The centre started through a major research process that involved more than 1000 researchers competing for more than 100 research proposals in a peer-review style approach. From therein, the institute delivered a number of capability-led workshops and user-led summits that involved a number of data science disciplines and leaders from all walks of data science roles. The result of the findings shaped how and what the centre devoted it’s time to in terms of research areas, and those areas now form the basis of all research at the institute.

In short, the chosen research areas were determined by where the most impact could be made through applied data science approaches.

Projects at the Institute

The institute has had many projects in the last couple of years and is currently involved in multiple projects across many areas, so it would be impossible to document everything that the institute Is working on. So, here’s a look at some of the most current projects that the researchers are involved in at the Institute.

The latest project is entitled Statistical Techniques for Engineering with Advanced Materials (STEAM). The project is a joint venture between the Alan Turing Institute, Imperial College London, Newcastle University, the University of Bath, the University of Exeter and King’s College London. The project bridges many areas of science including statistics and applied mathematics and involves physical experiments to investigate the material properties of 3D-printed stainless steel. The project also involves novel statistical methods to help provide an insight and analysis into the novel properties of the material.

Another new project involves using anomaly detection to combat fraud. This is a privately funded venture in conjunction with Accenture looking at anomalies in networks which are partly responsible for the annual loss of 5% of the global GDP caused by fraud and corruption. The project analyses similarity matrices between networks to find patterns in the data which vary from the expected behaviour of the matrix. It is a project with real-world importance and the implications of the research could find itself in fraud detection for credit cards and bank transactions, insurance claims, and money-laundering detection.

The next project that I’m going to mention is in the measurement and analysis of the world’s first (and largest) 3D printed steel bridge. The Institute has partnered with MX3D to provide the analysis and monitoring component of the project for a bridge that is to be installed in Amsterdam. The bridge is set to go across a canal and is going to be 12 metres in length. A team of structural engineers, mathematicians, computer scientists and statisticians from the institute are involved in the design and implementation of a vast sensor network to collect data on the strain, displacement and vibration of the structure. The team will also be involved in measuring the environmental factors surrounding air quality and temperature to measure and monitor the bridge’s health in real time. The data will also be inputted into a ‘digital twin’, which is a living computer model of the bridge, to provide insights into the effects that frequent usage has on its long-term structural integrity.

One project of commercial interest is in disaster management and is a joint venture between the Institute, Airbus, UK Space Agency, Catapult Satellite Applications and the EPSRC. The project involves creating an effective situation awareness technology from the myriad of data available from satellites, tweets, UAVs and in-situ sensors. Making use of all the data in a timely manner is an important protocol in disaster management and the team aims to create apps and dashboards to exploit recent advances in crowdsourcing, Bayesian classifier combination, deep learning, heterogeneous data fusion and scalable inferences to create more efficient disaster plans and responses. This work is currently being developed for use across Malaysia, Ethiopia and Kenya. Steve Reece, who heads up the project, explains “The preponderance of different kinds of data, from twitter to satellite imagery, is critical to mitigating and responding to natural disasters. Machine learning can make sense of this data and support both policy decisions, disaster planning and response. As natural disasters increase in frequency and ferocity, the need for scalable and rapid data analytics is becoming more important. Oxford and now the ATI are incredible environments that support the development of foundational disaster data science as well as encourage the opportunity to put theory into practise to help save lives. Our work with disaster responders in the Caribbean following the recent hurricanes is testament to the power of machine learning for social good.”

The final project of note is that on electric vehicle charging, which is funded by the Lloyd’s Register Foundation. The project looks to apply mathematical modelling principles to real-world data sets and to develop a decentralised scheduling approach for electric vehicle charging. The project will enable a central authority to oversee the network and ensure that vehicle owners can represent their preference towards electric vehicle charging levels and cost. The project is expected to be of great benefit to the National Grid and could help with making renewable energies compatible with the UK energy infrastructure.

Obviously, there are many more projects in the pipeline and many more are currently being undertaken. If you are interested in reading about some more of the projects at the Institute, then I would urge you to follow this link and discover more about the various research aspects of the Institute.

Sources:
Turing Institute- https://www.turing.ac.uk/
Alan Turing:The Enigma- http://www.turing.org.uk/
University of Cambridge Computer Laboratory: https://www.cl.cam.ac.uk/projects/raspberrypi/tutorials/turing-machine/one.html