Constructing Error Correcting Output Coding Classifiers based on weighted least squares

摘要

The object of this paper is to develop the Error Correcting Output Coding(ECOC) classifier. The ECOC classifier is one of Multiple Classifier Systems(MCSs), which prepare a large number of discriminant functions and then construct a classifier by combining these discriminant functions. We focus on a method for combining the discriminant functions and develop an extension of ECOC. Experiments on artificial and real data demonstrate that a much better performance can be obtained by the extended ECOC classifier.