论文部分内容阅读
传统的支持向量机是将分类问题转化成二次规划问题来解决的。针对传统的支持向量机算法及其变形算法忽略了训练集数据含有较大人为误差参与时其算法精度所存在的保障问题,提出了基于人为误差的支持向量机(artificial error—support vector machine以下称AE-SVM)的基本理论,并建立了AE-SVM的理论模型。该模型是C-SVM模型的改进和推广。
The traditional support vector machine is to solve the classification problem into the quadratic programming problem. Aiming at the traditional support vector machine algorithm and its deformation algorithm, the problem of guaranteeing the accuracy of the algorithm when the training set data contains larger human errors is neglected. An artificial error-support vector machine AE-SVM), and set up a theoretical model of AE-SVM. This model is an improvement and promotion of C-SVM model.