【作者】 门洪； 武玉杰； 李小英； 高艳春；

【Author】 Men Hong Wu Yujie Li Xiaoying Gao Yanchun (School of Automation Engineering Northeast Dianli University Jilin 132012 China)

【机构】 东北电力大学自动化工程学院；

【摘要】 本文采用支持向量机算法在UCI标准数据集上进行了分类实验。结果表明:对于训练样本较少的数据集,采用结构简单的学习函数,而对于训练样本较多的数据集,结构简单的学习函数会降低机器学习的推广能力;通过适当增大惩罚因子的值,可以提高学习机的推广能力;采用交叉验证法选择核参数可以提高分类精度;支持向量机分类算法的识别准确率优于神经网络算法。更多还原

【Abstract】 The classification experiments on the UCI database are deployed with support vector machine method. We can make a conclusion that,as for the dataset with less training samples,simple structure learning function will be applied.In contrast,the learning function with slightly simple structure will reduce the generalization ability.Increasing the value of penalty factor appropriately,the generalization ability of the learning machine can be improved.Using cross-validation method to choose parameter values can improve the classification accuracy. Experimental results show that the support vector machine method is superior to the neural network algorithm.更多还原