Optimal Extension of Error Correcting Output Codes

Sergio Escalera, Oriol Pujol, Petia RadevaCCIA 2006 · OpenReview

Abstract

Error correcting output codes (ECOC) represent a successful extension of binary classifiers to address the multiclass problem. In this paper, we propose a novel technique called ECOC-ONE (Optimal Node Embedding) to improve an initial ECOC configuration defining a strategy to create new dichotomies and im-prove optimally the performance. The process of searching for new dichotomies is guided by the confusion matrices over two exclusive training subsets. A weighted methodology is proposed to take into account the different relevance between dichotomies. We validate our extension technique on well-known UCI databases. The results show significant improvement to the traditional coding techniques with far few extra cost.
BibTeX
@inproceedings{escalera2006optimal,
  author = {Sergio Escalera and Oriol Pujol and Petia Radeva},
  title = {{Optimal Extension of Error Correcting Output Codes}},
  booktitle = {CCIA},
  year = {2006},
  url = {https://openreview.net/forum?id=D7jG83iTz3},
}
Figure