Towards Learning to Rank in Description Logics
Nicola Fanizzi, Claudia d’Amato, Floriana Esposito · Frontiers in artificial intelligence and applications · 2010
In the context of knowledge bases expressed in Description Logics, a method for learning functions that can predict the ranking of resources encoding some preference criteria implicitly encoded through examples of rated individuals. The method relies on a kernelized version of the PERCEPTRON RANKING algorithm which is suitable for batch but also online problem settings.