Você está aqui: Entrada News & Events INESC TEC receives award for paper on data mining techniques to predict malfunctions

INESC TEC receives award for paper on data mining techniques to predict malfunctions

The paper “Failure Prediction - an Application in the Railway Industry”, written by Pedro Mota Pereira, Rita Ribeiro and João Gama, researchers at INESC TEC’s Laboratory of Artificial Intelligence and Decision Support (LIAAD), has received the Carl H. Smith Student Paper Award at the conference Discovery Science 2014.

The paper addresses the theme of exploring data to predict malfunctions, using as case study the doors of suburban trains operating in England. The researchers used data mining techniques to verify if it was possible to detect malfunctions in the doors at an early stage. “Throughout the work, we ended up demonstrating that applying post-processing to the results in the first classification stage was an adequate way of predicting malfunctions in systems such as automatic doors in trains”, Pedro Mota Pereira explains.

The awarded work is the result of research that is currently being developed by Pedro Mota Pereira as part of his Master’s Degree in data analysis at the Faculty of Economics of the University of Porto.

This is the second time that a LIAAD researcher receives the Carl H. Smith award. In 2006, the award went to Rita Ribeiro, co-author of the paper that was now awarded.

The Carl H. Smith award exists since 2005 and it is given annually at the Discovery Science conference. The researchers received their award during the conference, which took place between 8 and 10 October, in Slovenia.

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

Quem Somos

O Connect INESC TEC junta os antigos colaboradores do INESC TEC. Pretendemos manter ligações fortes com os nossos antigos colaboradores, que podem aqui receber notícias sobre o Laboratório Associado e manter contacto com os seus antigos colegas.

Cadê Você?

Ler mais