Multi-class learning and error-correcting code sensitivity
暂无分享,去创建一个
Properties of optimal error-correcting codes for decomposing a multi-class problem into a set of two-class problems are discussed. After analysing the effect of the code matrix on individual classification errors and on errors due to combining, the authors propose a modification that is less sensitive to code word selection.
[1] Terrence J. Sejnowski,et al. Parallel Networks that Learn to Pronounce English Text , 1987, Complex Syst..
[2] Catherine Blake,et al. UCI Repository of machine learning databases , 1998 .
[3] Thomas G. Dietterich,et al. Solving Multiclass Learning Problems via Error-Correcting Output Codes , 1994, J. Artif. Intell. Res..
[4] W. W. Peterson,et al. Error-Correcting Codes. , 1962 .