【摘要】 建立了一种基于支持向量机的地表水环境质量分类模型,并将其用于浙江省主要市界交界面的地表水环境质量分类。该模型采用径向基核函数,以一对多方式实现多分类。分别以网格搜索、粒子群优化和遗传算法三种优化方法对支持向量机的控制参数进行寻优。实验表明,采用网格搜索法确定支持向量机控制参数可以得到最好的水质分类结果,分类准确率可达到82%,由此证明以支持向量机对水质进行分类是可行的。更多还原

【Abstract】 This paper presents a classification model of environmental quality for surface water based on support vector machine(SVM). The model is used to classify the surface water quality of the interface between the major cities in Zhejiang province. The model uses one-against-all multi-class SVM classifier with radial basis kernel function. Grid-search, particle swarm optimization and genetic algorithm were respectively used to optimize the control parameters of the SVM. Experimental result shows that the grid search method can achieve the best classification accuracy which reaches up to 82%. The result indicates that the proposed method is feasible.更多还原