论文部分内容阅读
支持向量机以统计学习理论为基础,采用结构风险最小化准则,将学习问题转化为一个凸二次规划问题,能够得到全局最优解,适合解决小样本、非线性分类及回归问题。根据水资源安全的内涵,筛选出具有代表性的指标,组成水资源安全评价指标体系。建立了基于支持向量机的水资源安全评价模型,将安全标准划分为良好、安全、临界、不安全、危险5个等级。根据水资源安全评价标准及所属评价等级值,随机生成样本集,180个样本作为训练样本,构造了5个两类支持向量分类器,20个样本作为检验样本,检验样本分类全部正确。将模型应用于山西省11个城市的水资源安全评价,结果表明,该方法是有效、可行的。
Based on statistical learning theory, the support vector machine (SVM) adopts the minimization criterion of structural risk to transform the learning problem into a convex quadratic programming problem, which can obtain the global optimal solution and is suitable for solving small samples, nonlinear classification and regression problems. According to the connotation of water resources security, selected the representative indicators, composed of water resources safety evaluation index system. The water resources safety evaluation model based on support vector machine is established. The safety standards are divided into five levels: good, safe, critical, unsafe and dangerous. According to the standard of water resources safety evaluation and the corresponding evaluation grade value, a sample set is randomly generated and 180 samples are used as training samples. Five types of support vector classifiers are constructed. Twenty samples are used as test samples, and the test sample classification is all correct. Applying the model to the water resources safety assessment of 11 cities in Shanxi Province, the results show that this method is effective and feasible.