Combining the new measure with the forward regression orthogonal least square (OLS), not only the parameters of the classification hyperplane, but also the important input nodes can be obtaind.

• 提出一种基于输入集分类函数的新的距离度量方法,它与前传回归的正交最小二乘法相结合,不仅可以学习分类超平面的参数,而且可以选择重要的输入节点。