Consumer Credit Award

The retail credit domain is characterised by data sets which are large in terms of number of cases, with billions of card transactions per year, and in terms of number of variables, with some 70-80 items recorded for each card transaction, expanded to hundreds or even thousands in fraud detection applications.

It was “big data” before the term was invented. It’s an area which presents many novel mathematical, statistical, and machine learning challenges, requiring the development of new methodology.

The Consumer Credit Research Group at Imperial has contributed significantly to these developments, creating new and improved methods of scorecard evaluation, new tools for fraud detection, strategies for coping with selection bias, and in other areas.

The group received the Credit Collections and Risk award for contributions to the credit industry in 2012, the first academic research group to receive this award.