论文部分内容阅读
为了提高孪生支持向量机的泛化能力,提出一种新的孪生大间隔分布机算法,以增加间隔分布对于训练模型的影响.理论研究表明,间隔分布对于模型的泛化性能有着非常重要的影响.该算法在标准孪生支持向量机优化目标函数上增加了间隔分布的影响,间隔分布通过一阶和二阶数据统计特征来体现.在标准数据集上的实验结果表明,所提出的算法比SVM、TWSVM、TBSVM算法的分类精确度更高.
In order to improve the generalization ability of twin SVM, this paper proposes a new twin big-interval splitter algorithm to increase the influence of interval distribution on the training model.The theoretical study shows that the interval distribution has a very important influence on the generalization performance of the model The algorithm increases the influence of the interval distribution on the optimized objective function of the standard twin SVM, and the interval distribution is represented by the first-order and the second-order statistics.Experimental results on the standard dataset show that the proposed algorithm is better than SVM , TWSVM, TBSVM algorithm classification accuracy higher.