基于类间最近邻支持向量信息测度排序的快速分类算法研究

A Fast Support Vector Classification Algorithm Based on the Sort of Nearest Neighborhood Information Measure

提出了基于特征空间中最近邻类间支持向量信息测度排序的快速支持向量机分类算法,对于训练样本首先进行最近邻类间支持向量信息测度升序排列处理;然后根据排序的结果选择最优的训练样本子空间,在选择的样本子空间内采用乘性规则直接求取Lagrange因子,而不是传统的二次优化方法;最后加入附加剩余样本进行交叉验证处理,直到算法满足收敛性准则。各种分类实验结果表明,该算法具有非常良好的性能,特别是在训练样本庞大,支持向量数量较多的情况下,能够较大幅度地减少计算复杂度,提高分类速度。

To improve the training speed performance of large-scale support vector machine(SVM), a fast algorithm is proposed in this paper by exploiting the geometric distribution of support vector in feature space. A support vector information measure definition based on the nearest inter-classes distance is set up and a sort process is presented. Then a reduced number of sample subspace is extracted for support vector training. In addition, instead of the traditional quadratic programming, multiplicative update is used to solve Lagrange multiplier in optimizing the solution of support vector. The samples of rest are used for cross validating till the algorithm is convergence. Experimental results demonstrate that this method has better performance and has overcome the flaw of standard SVM. This algorithm could greatly reduce the computational load and increase the speed of training, especially in the case of large number of training samples.