摘要
支持向量机的训练速度慢,制约了它的发展和推广应用。Suykens提出了一种新的支持向量机方法——最小二乘支持向量机。最小二乘支持向量机是支持向量机的发展和改进,它采用等式约束替代不等式约束,求解速度大大加快。将其用于大坝的渗流监测中,并与传统的支持向量机进行了比较,结果显示二者的预测效果都比较好,但是最小二乘支持向量机的训练效率比支持向量机要高。
Training slow of SVM restrict its development and application. Suykens presents a new method of support vector machine - least squares support vector machine. Least squares support vector machine is the development and improvement of support vector machines. Lssvm use equality constraints to alternative inequality constraints. So the speed of solution accelerates greatly. In this paper it is used in dam seepage monitoring and compared with the traditional support vector machine. The results show that the predicting effect of the two methods is good relatively. But the training efficient of least squares SVM is higher than support vector machine.
关键词
大坝渗流
支持向量机
最小二乘支持向量机
dam seepage
support vector machine
least squares support vector machine
