Multiple Classifier Fusion Using k-Nearest Localized Templates
This paper presents a method for combining classifiers that uses k-nearest localized templates. The localized templates are estimated from a training set using C-means clustering algorithm, and matched to the decision profile of a new incoming sample by a similarity measure. The sample is assigned to the class which is most frequently represented among the k most similar templates. The appropriate value of k is determined according to the characteristics of the given data set. Experimental results on real and artificial data sets show that the proposed method performs better than the conventional fusion methods.
KeywordsClassifier fusion Decision templates C-means clustering
Unable to display preview. Download preview PDF.
- 14.Min, J.-K., Hong, J.-H., Cho, S.-B.: Effective Fingerprint Classification by Localized Models of Support Vector Machines. In: Zhang, D., Jain, A.K. (eds.) ICB 2006. LNCS, vol. 3832, pp. 287–293. Springer, Heidelberg (2005)Google Scholar