论文部分内容阅读
<正>The computational cost of support vector regression in the training phase is O(N~3),which is very expensive for a large scale problem.In addition,the solution of support vector regression is of parsimoniousness, which has relation to a part of th