1) Minimum classification error
最小分类错误
1.
Recently, many researchers pay more attention to an important classific ation method, which is called the discriminative learning method based on minimum classification error criterion.
介绍了一种近年来正被广泛重视的模式识别方法——基于最小分类错误准则的判别学习方法 ,讨论了该方法的基本原理及其可能的应用领
2.
One most popular technique is minimum classification error training.
隐马尔可夫模型的参数估计是研究的重点话题,其中一个主要的方法是最小分类错误。
2) MCE
最小分类错误
1.
A Minimum Classification Error(MCE) criterion based sub-words weighting parameters estimation algorithm is proposed in which the sub-word weighting parameters are derived by the MCE training.
本文提出了一种基于最小分类错误准则(MCE)的子词权重参数估计算法,通过MCE训练得到子词的权重系数。
3) ｍｉｎｉｍｕｍ ｃｌａｓｓｉｆｉｃａｔｉｏｎ ｅｒｒｏｒ
最小错误分类
4) minimum classification error criterion
最小分类错误准则
5) misclassification minimization
错误分离最小化
1.
Approximation algorithm of misclassification minimization;
错误分离最小化的一个新的近似算法
6) minimum classification error
最小分类误差
1.
Based on MultiBoost-a classification ensemble algorithm,this paper promotes the algorithm that can obtain the minimum classification error of MultiBoost by increment cross-validity so that under the specified size T of MultiBoost ensemble,it can find out the composite classifier with the minimum classification error.
基于MultiBoost分类组装技术,提出了一种用增量交叉验证技术求MultiBoost最小分类误差的算法,以使之在指定分类器数量T的范围内找出具有最小分类误差的合成分类器。
补充资料：错误分类偏倚
错误分类偏倚
错误分类是指将一个调查对象，一个数值或一个特征错误地分到不是它所属的类别中。
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条