INESC TEC paper receives “Best Student Paper Award”

The paper “oAdaBoost – An AdaBoost Variant for Ordinal Classification”, written by João Costa (former collaborator) and Jaime Cardoso, of INESC TEC’s Centre for Telecommunications and Multimedia (CTM), received a “Best Student Paper Award” at the 2015 International Conference on Pattern Recognition Applications and Methods (ICPRAM).

This work studies the potential of ensemble methods to classify ordinal data, and generalises a recently proposed method following this paradigm. Ordinal data classification basically consists of predicting ordinal categories and can be applied in several real life domains, where human evaluation plays an important role. For instance, it can help choose a song, an item of clothing or a meal at a restaurant – situations that require the user to compare or classify the most suitable item. At a more complex level, ordinal data classification can be useful in stock trading support systems in which it is possible to predict the possibility of buying, keeping or selling stocks.

This paper was written as part of João Costa’s Master in Informatics and Computer Engineering at the Faculty of Engineering of the University of Porto (MIEIC/FEUP), advised by Jaime Cardoso.

The ICPRAM 2015 took place in Lisbon between 10 and 12 January. This forum gathers researchers and professionals in the theory and application of pattern recognition.

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