【摘要】 支撑矢量机为小样本模式识别提供了一新的途径 ,但其支撑矢量的选择相当困难 ,也成为其应用的瓶颈问题 .对此 ,本文提出了一种能够预先选取支撑矢量的方法———中心距离比值法 .该方法在不影响支撑矢量机的分类能力情况下 ,大大地减少了训练样本 ,提高了支撑矢量机的训练速度 .文中给出的仿真实验结果也验证了该方法的有效性和可行性 .类似的结果在国内外还未见报导更多还原

【Abstract】 A new method called center distance ratio which is able to extract support vectors from given training examples is presened for support vector algorithm.The method greatly reduces the training samples and so improves the speed of support vector machine,while the ability of support vector machine to classification is unaffected.Our experimental results show remarkable improvement of speed support vector to support our idea.更多还原