Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
Article

LIAAD/INESC TEC wins 1st place in finances and data mining competition

A team with members from INESC TEC’s Laboratory of Artificial Intelligence and Decision Support (LIAAD) and from the Faculty of Economics of the University of Porto (FEP) has recently won a competition on credit rating as part of the conference BRICS-CCI & CBIC 2013 - Countries Congress on Computational Intelligence, which took place between 3 June and 19 August. The results were announced during the conference’s closing session in Brazil on 10 September.

10th October 2013

The competition, open to academy and industry (financial, rating and analytics), focused on the application of credit scoring in the retail banking sector,  and its goal was to promote innovative solutions  to deal with the temporal  degradation of credit rating models. During this period, the contestants developed and tested their solutions interactively and in real time via web.

The Portuguese team, which included João Gama (a researcher at LIAAD and professor at FEP), Maria do Carmo Sousa (PhD student at FEP) and Manuel Silva Gonçalves (FEP alumnus), proposed a new two-phase tool to deal with the temporal degradation of credit scoring using as case study a set of clients with credit cards from a Brazilian institution. The prediction made by the Portuguese researchers took into consideration not only the time evolution, but also the evolution of data following changes in the model’s population, in the economy and in the market. The methodology presented produced encouraging results for a time-horizon of one year. The researchers believe that the model can be used successfully in other risk assessment applications.

Four finalists were selected out of 60 teams registered. Other than the LIAAD/FEP team, the other finalists were the University of Technology and Economics in Budapest (Hungary), StatSoft (Norway) and the Federal University of Rio de Janeiro (Brazil). Following the competition, the Portuguese researchers published a paper titled “A two-stage model for dealing with temporal degradation of credit scoring”, integrated in the proceedings of BRICS-CCI & CBIC 2013.

The INESC TEC researchers mentioned in this article are associated with the following partner institutions: FEP.

INESC TEC, October 2013