一种改进的样例约简支持向量机

An improved instance reduction support vector machine

在以前工作的基础上,提出了一种改进的样例约简支持向量机,利用相容粗糙集方法求属性约简的边界域,并从中选择样例作为候选支持向量训练支持向量机.该方法的特点是可同时对属性和样例进行约简.实验结果证实了这种方法的有效性,能有效地减少存储空间和执行时间.

Based on the previous work,an improved instance reduction support vector machine was proposed in this paper. By employing tolerance rough set technique, the boundary region of the reduct is calculated and the candidate support vectors used for training support vector machine are selected from this region. Simultaneous calculations of attribute reduction and instance reduction characterize our method. The experimental results show that the proposed method is effective and can efficiently reduce the computational complexities of both time and space.