Error Correcting Output Coding is a well established technique to decompose a multi-class classification problem into a set of two-class problems. However, a point not yet considered in the research is how to apply this method to a cost- sensitive classification that represents a significant aspect in many real problems. In this paper we propose a novel method for building cost-sensitive ECOC multi-class classifiers. Starting from the cost matrix for the multi-class problem and from the code matrix employed, a cost matrix is extracted for each of the binary subproblems induced by the coding matrix. As a consequence, it is possible to tune the single two-class classifier according to the cost matrix obtained and achieve an output from all the dichotomizers which takes into account the requirements of the original multi-class cost matrix. To evaluate the effectiveness of the method, a large number of tests has been performed on real data sets. The first experimental results show that the proposed approach is suitable for future developments in cost-sensitive application.
A Method for Designing Cost sensitive ECOC
MARROCCO, Claudio;TORTORELLA, Francesco
2004
Abstract
Error Correcting Output Coding is a well established technique to decompose a multi-class classification problem into a set of two-class problems. However, a point not yet considered in the research is how to apply this method to a cost- sensitive classification that represents a significant aspect in many real problems. In this paper we propose a novel method for building cost-sensitive ECOC multi-class classifiers. Starting from the cost matrix for the multi-class problem and from the code matrix employed, a cost matrix is extracted for each of the binary subproblems induced by the coding matrix. As a consequence, it is possible to tune the single two-class classifier according to the cost matrix obtained and achieve an output from all the dichotomizers which takes into account the requirements of the original multi-class cost matrix. To evaluate the effectiveness of the method, a large number of tests has been performed on real data sets. The first experimental results show that the proposed approach is suitable for future developments in cost-sensitive application.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
