Probabilistic Multi-Label Learning for Medical Data
Damien Zufferey · ArODES (HES-SO (https://www.hes-so.ch/)) · 2014
We report on a probabilistic approach for the classification of chronically ill patients. We rely on multi-label learning for its ability to represent in a natural way classification problems involvingcoexistenceofdiseases.Weuseapublicclinicaldatabase for the evaluation of our proposed algorithm. Preliminary results show the benefits of our approach.