Abstract: A new support vector machine based on sample probabilistic estimated value is presented in this paper.The sample probabilistic estimated value and the distance between the sample and the super plane are defined，then this new support vector machine is formed.It is fit for both linear and nonlinear conditions.Experiments show that the support vector machine has better results for classification and less running time than other SVMs.

Key words: classification, support vector machines, data sample, probabilistic SVMs, probabilistic estimated value

摘要： 提出一个新的基于样本点概率估计的支持向量机，通过定义相应样本数据点的概率估计值，以及相应的数据样本点到超平面的距离，来形成新的线性和非线性情况下的支持向量机。最后通过实验证明，在数据集的训练上，新的支持向量机比以往传统的支持向量机有更好的分类性能，并且缩短了支持向量机数据样本的训练时间。

关键词: 分类, 支持向量机, 数据样本, 概率支持向量机, 概率估计值