装备费用预测中的混合核函数支持向量机

在支持向量机预测建模中,核函数用来将低维特征空间中的非线性问题映射为高维特征空间中的线性问题.核函数的特征对于支持向量机的学习和预测都有很重要的影响.考虑到两种典型核函数—全局核(多项式核函数)和局部核(RBF核函数)在拟合与泛化方面的特性,采用了一种基于混合核函数的支持向量机方法用于预测建模.为了评价不同核函数的建模效果、得到更好的预测性能,采用遗传算法自适应进化支持向量机模型的各项参数,并将其应用于装备费用预测的实际问题中.实际计算表明采用混合核函数的支持向量机较单一核函数时有更好的预测性能,可以作为一种有效的预测建模方法在装备管理中推广应用.

Support Vector Machine with Composite Kernel Functions in Equipment Cost Forecasting

Kernel functions,in Support Vector Machine(SVM),are applied to map a nonlinear model into a higher dimensional feature space where a linear learning is adopted.The characteristics of kernel functions have great impacts on learning and predictive results of SVM.Considering the characteristics for fitting and generalization of two kinds of typical kernel functions-global kernel(polynomial kernel) and local kernel(RBF kernel),a new kind of SVM modeling method based on composite kernel functions was adopted.In order to evaluate the reasonable fitness of kernel functions,genetic algorithm was used to adaptively evolve SVM to obtain the best forecasting performance.Experiments in the equipment cost forecasting demonstrate that the SVM with composite kernel functions has the better performance than with a single kernel function and can be popularized as an effective method.