最小二乘支持向量机的参数优选方法及应用

Method for selecting parameters of least squares support vector machines and application

支持向量机是一种新型的学习方法,该方法以结构风险最小化原则取代传统机器学习中的经验风险最小化原则,在小样本的机器学习中显示出了优异的性能.传统的支持向量机是解凸二次规划问题,而最小二乘支持向量机是解等式线性方程,显得尤为方便,通过建立适当的性能指标,用遗传算法优化最小二乘支持向量机的有关参数,并在非线性经济系统中应用.用最小二乘支持向量机对非线性经济系统进行预测并与其它方法的预测结果比较,结果证明,该模型的预测精确度是令人满意的,文中提出的方法是可行的.

Support vector machine （SVM） is a kind of novel learning method, which is based on structural risk minimization principle, unlike traditional machine learning which is based on empirical risk minimization principle. SVM has shown powerful ability in learning with limited samples. But in SVM, the solution of the problem is characterized by a convex quadratic programming problem. In least squares SVM （LS-SVM）, a least squares cost function is proposed so as to obtain a linear set of equations in dual space. Through genetic algorithm, parameters optimization can be solved. The model of LS-SVM is then used to forecast a nonlinear macroeconomic system. It is shown that the LS-SVM based on parameters optimization by genetic algorithm is simple and effective.