一种改进的基于支持向量机的概率密度估计方法

统计学习理论是针对小样本数据而提出的一套理论，支持向量机方法可用于解决有限样本情况下的概率密度估计问题，该种方法与Parzen窗的精度等级类似，同时又具有Parzen窗方法所不具备的稀疏解，但应用过程中耗时较多。本文提出了一种基于支持向量机的改进密度估计方法，分析了其原理，通过仿真实验的对比，改进后的密度估计方法与改进前相比，估计精度的水平保持不变，训练后的支持向量数目不变，对训练样本的训练时间大幅缩短。由仿真结果可知，本文提出的对支持向量机法估计概率密度的改进是可行的。

An Improved Algorithm Based on SVM for Probability Density Estimate

Statistical learning theory is a theory based on small sample data, which has been considered to be an important complement and development of traditional statistics. The support vector machine （SVM）is selected to estimate the probability density for small sample data. Compared with Parzen window, the results of this method have similar quality and sparse solutions except cost a long time. To enhance the computational efficiency of SVM algorithm, an improved algorithm is proposed. Training the same samples, the improved algorithm shortens the train time while the precise of estimation is not changed.