不确定性支持向量分类预警算法

支持向量分类从理论上可以保证好的预警外推能力。但历史警度的确定是一个十分棘手的问题。论文提出了不确定性支持向量分类预警方法。将支持向量分类预警问题转化为各个历史样本的惩罚系数的合理变化,从而大大减少了约束的个数,体现了专家决策在预警系统的作用。不仅实现了专家意见的综合,而且是对SVM理论本身的拓广。证明了模糊支持向量机是不确定性支持向量分类的特例,从而给出了模糊支持向量机的确切含义。数据试验表明,未确知支持向量分类预警方法具有一定的实际应用价值。

Uncertainty Support Vector Classification for Early-Warning

Support Vector Classification has shown great performance in practice as a two class classification methodologyHowever,it is required that every input must be exactly assigned to one of these two classes without any uncertainty. In this paper,we propose USVC early-warning algorithm on expert advices,which is able to deal with the training data with uncertaintyThe application field of our algorithms is,wider than the SFVM method,although uncertainty is also considered thereMoreover,early warning of grain security is solved by USVC algorithm.