The method uses a self-organizing map to obtain the class label for each training sample and enhanced Fisher linear discriminant ( EFM) to find the optimal projection for pattern classification, and a Gaussian distribution to model the class-conditional density function of the projected samples for each class. 该方法首先使用自组织映射网络为每个训练样本确立类别标签,然后用改进的Fisher线性判别模型对所有样本进行投影以尽可能拉大各类之间的距离,最后使用高斯分布对每类样本进行建模。
Uniformly convergent rate of N.N estimate of the ■~ ( 2) class density function under general kernel ■~((2))类密度函数在一般核下最近邻估计的一致收敛速度
In the algorithm, the class conditional probability density function was defined as the reciprocal of the distance between sample X and the cluster center. 在概率聚类算法中,类条件概率密度函数定义为样本X到该类聚类中心之间距离的倒数。