Paper Info
Title
A Method for Designing Cost-Sensitive ECOC
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.
Year
DOI
Venue
2004
10.1007/978-3-540-25966-4_20
Lecture Notes in Computer Science
Keywords
Field
DocType
multi class classification
Signal processing,Data set,Cost matrix,Matrix (mathematics),Computer science,Algorithm,Coding (social sciences),Error detection and correction,Classifier (linguistics),Binary number
Conference
Volume
ISSN
Citations 
3077
0302-9743
0
PageRank 
References 
Authors
0.34
10
2
Name
Order
Citations
PageRank
Claudio Marrocco18417.53
Francesco Tortorella237043.39