Fusion of Classifiers based on a Novel 2-Stage Model

View/ Open
File version
Accepted Manuscript (AM)
Author(s)
Nguyen, TT
Liew, AWC
Tran, MT
Thu Thuy Nguyen, T
Nguyen, MP
Griffith University Author(s)
Year published
2014
Metadata
Show full item recordAbstract
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 ...
View more >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.
View less >
View more >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.
View less >
Conference Title
Communications in Computer and Information Science
Volume
481
Publisher URI
Copyright Statement
© 2014 Springer Berlin/Heidelberg. This is the author-manuscript version of this paper. Reproduced in accordance with the copyright policy of the publisher. The original publication is available at www.springerlink.com
Subject
Expert Systems