摘要
感知机只能解决线性可分问题。支持向量机中的L2范数软边缘算法可以将线性不可分问题转化为线性可分问题。基于这一事实,提出一种基于L2范数的软核感知机(SoftKernelPerceptron,SKP),将感知机算法直接用于求解L2范数软边缘算法决定的线性可分问题。通过使用核技巧,得到一种普适的非线性分类方法。实际数据库的测试结果表明,SKP算法能够有效地解决非线性问题,并且继承了感知机运算简单速度快的优点。
The perceptron can only solve linearly separable problems.The L2 norm soft margin algorithms in SVMs can change each linearly inseparable problem into a separable one.Based on this fact,we propose a Soft Kernel Perceptron(SKP) in terms of L2 norm,in which the regular perceptron is directly employed to solve the linearly separable problem determined by L: norm soft margin algorithms.By using kernel technique,a general method for solving nonlinear classifications is obtained.The experiments on some real datasets demonstrate that the proposed SKP can solve nonlinear classification efficiently.Moreover,it has the original advantage of simple calculation and fast speed.
出处
《计算机工程与应用》
CSCD
北大核心
2007年第2期91-94,共4页
Computer Engineering and Applications
关键词
感知机
核函数
线性可分
线性不可分
L2范数
perceptron
kernel function
linearly separable
linearly inseparable
L2 norm