Fusion of Classifiers based on a Novel 2-Stage Model

Abstract

The paper introduces a novel 2-Stage model for multi-classifier system. Instead of gathering posterior probabilities resulted from base classifiers into Level1 data like in the original 2-Stage model, here we separate data in K Level1 matrices corresponding to the K base classifiers. These data matrices, in turn, are classified in sequence by a new classifier at the second stage to generate Level2 data. Next, Weight Matrix is proposed to combine Level2 data and predict label of observations in test set. Experimental results on CLEF2009 medical image database demonstrate the benefit of our model in comparison with several ensemble learning models.