A classifier fusion method based on classifier accuracy
Wenxing Li, Jian Hou, Lizhi Yin · 2014
Classifier fusion methods are usually used to combine multiple classification decisions and generate better classification results than any single classifier. In order to improve object classification accuracy, it is a common method to assign weights to classifiers based on their importance in a multiple decision system. In this paper we put forward a method to weight different classifiers in classifier fusion. With SVM as the classifier, we use cross-validation to estimate the accuracy of classifiers and therefore the importance of classifiers in classifier fusion. In the next step we test different weighting methods based on the classifier accuracy and select the best performing one. In experiments on three diverse datasets, our method performs better than average fusion and the best single classifier, and also comparable to the state-of-the-art methods in literature.