.

Machine learning for visual analytics

Aston research is helping businesses better understand large and complex datasets through visualisation tools

Visualisation of datasets

Visual analytics is a powerful approach to understanding large and complex datasets that makes information accessible to nonstatistically trained users.

The Non-linearity and Complexity Research Group (NCRG) at Aston has developed several fundamental algorithms and extended their applicability with interactive software tools such as Netlab and DVMS

Netlab and DVMS Software

There have been more than 40,000 downloads of Netlab by the academic and business communities worldwide. Visual analytics and the DVMS software have played an important role in improving business performance and collaborative industrial research projects to understand data, detect outliers and select important features.

The software is used by industrial partners including Pfizer and the Defence Science and Technology Laboratory, who have incorporated the data visualisation algorithms of the NCRG into their business activities. The group at Aston University has developed an interactive visualisation tool for Pfizer’s research chemists and biologists to interpret and analyse screening results such as biological activity and toxicity. A prototype information visualisation system related to integrating spatio-temporal and network analysis incorporating measures of uncertainty was delivered to the Defence Science and Technology Laboratory

The visualisation algorithms have also been integrated into a commercial tool by Integrated Geochemical Interpretation Ltd. This is a petroleum geochemical consultancy company which operates world-wide and sells 'p:IGI', which is a software product, for geochemical interpretation and basin modelling.

Visual analytics has played an important role in the development of new products, such as the Body Volume Index system

The motor company, 'WheelRight' installed three complete systems which are used each day by a range of vehicles. At a municipal bus depot, a fleet of 80 buses with over 2,000 tyre pressures, are measured each week. For a HGV fleet operator, this increases to over 14,000 and at the entrance to a technology site, over 10,000 tyre pressures are recorded.

The company has automatically recorded over a million tyre pressures in the last year from these three sites alone. Taking this number of tyre pressures manually would be impractical and very expensive, with estimated labour costs exceeding £500k per annum. Because the software is opensource, it is used to educate and train skilled people in institutions in the UK and internationally.

Staff Profile

Staff Profile

Professor Ian Nabney

Non-linearity and Complexity Research Group

"My research spans both the theory and applications of neural networks and other pattern recognition techniques, with a special focus on data visualisation and probabilistic modelling"

 

Staff Profile

Staff Profile

Professor David Lowe

Non-linearity and Complexity Research Group

David has worked alongside Professor Ian Nabney on the development of NetLab software