基于多尺度最小二乘支持向量机的舰船备件器材多类分类

针对远海任务舰船备件器材的分类管理，通过运用支持向量机理论，充分发挥多尺度核在非线性分类中的优势，借助最小二乘原理，构建了多尺度最小二乘支持向量机学习模型。在实际运用中，通过选用高斯径向函数作为多尺度核函数，以训练样本数据分布的离散系数作为核函数宽度参数取值依据，采取ECOC方法建立了多类分类模型，实例计算表明，该方法对远海任务舰船备件器材的分类是有效、可行的。

Characteristics Simplicity of Gene Batum Based on PCA and KPCA

In view of the classification management problems of pelagic task ship spare parts equipment, based on the support vector machines theory, making full use of the advantages of the multi-scale kernel in nonlinear classification, with the aid of the least squares principle, this paper establishes a multi-scale model of least squares support vector machine. In practical application, by selecting Gauss radical function as the multi-scale kernel function, taking the discrete coefficient of the training samples data distribution as the width parameter value reference of kernel function, the multiclass classification model is established by ECOC method. The result indi-cates that this method is effective and feasible for the classification of pelagic task ship spare parts equipment.