Presentation | 2005/3/23 Constructing error correcting codes based on estimation error of a posteriori probabilities for WLS-ECOC Nobuhiko YAMAGUCHI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The object of this paper is to construct error correcting codes for Error Correcting Output Coding(ECOC) classifiers. 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. In this paper, we focus on ECOC combination strategy based on Weighted Least Squares(WLS-ECOC) and propose a new construction of error correcting codes for WLS-ECOC. In this construction, we define the criteria of the codes as the estimation error of a posteriori probabilities and construct the codes based on the minimization of the estimation error. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | ECOC / error correcting codes / neural networks / pattern classification |
Paper # | NC2004-220 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 2005/3/23(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Constructing error correcting codes based on estimation error of a posteriori probabilities for WLS-ECOC |
Sub Title (in English) | |
Keyword(1) | ECOC |
Keyword(2) | error correcting codes |
Keyword(3) | neural networks |
Keyword(4) | pattern classification |
1st Author's Name | Nobuhiko YAMAGUCHI |
1st Author's Affiliation | Faculty of Science and Engineering, Saga University() |
Date | 2005/3/23 |
Paper # | NC2004-220 |
Volume (vol) | vol.104 |
Number (no) | 760 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |