改进型支持向量机在水质分类中的应用研究

Research on the Application of Supportive Vector Machine in the Classification of Water Quality

文章分别使用BP、RBF等神经网络和支持向量机等非线性方法对相同的水质数据建立分类模型。使用支持向量分类机建立水质分类模型过程中,选用RBF核函数,结合归一、降维等数据预处理手段,利用网格搜索算法对参数进行寻优,得出水质分类模型。实验结果证明在非线性方法中,采用支持向量机并结合相应的数据预处理手段这种方案得出的分类准确率更高,更加具有推广性。

My paper intends to build a model based on the application of artificial neural networks such as BP, RBF and non-linear method such as supportive vector machine in classifying the data on the same water quality. In such a process, using supportive vector machine, adopted radial basic function （RBF）, methodologies such as normalization, dimension reduction, and grid search algorithm to get optimization out of relevant parameter to classify the water quality, the results of my experiment suggest that among non-linear methods, combining the use of supportive vector machine with the relevant pre-processing data methods has proved more accurate in the classification, thus making it worth further promotion.