摘要
内点法是近年来发展起来的求解线性规划和二次规划的一种新方法,它有许多优点。支撑向量机问题中的二次规划问题有它特殊的一些性质,如它是一个凸二次规划,约束比较少而且都是线性约束,它的矩阵Q对称半正定且稠密。该文探索了用内点算法求解支撑向量机中这种特殊的二次规划问题,并给出了数值实验,表明这是一种好的求解支撑向量机的方法。
Interior point algorithm, which possesses a lot of advantages, is a new method for solving linear programming and quadratic programming. The quadratic programming in support vector machine has its particular properties, such as convex quadratic programming, a little and linear constraints, symmetric positive semi- definite and dense Q - matrix. In this paper, solving the quadratic programming in SVM by interior point algorithm is investigated. The numerical experiments demonstrate its usefulness.
出处
《西安邮电学院学报》
2005年第3期107-110,共4页
Journal of Xi'an Institute of Posts and Telecommunications
基金
西安邮电学院中青年科研基金项目资助
关键词
支撑向量机
内点算法
二次规划
support vector machine
interior point algorithm
quadratic programming
