基于支持向量机的小流域水蚀预报模型研究

 土壤侵蚀过程复杂,很难直接应用土壤侵蚀预报方程进行定量计算。作为一种新的机器学习算法,支持向量机在样本有限的情况下,采用结构风险最小化准则,把学习问题转化为一个二次规划问题,从而得到唯一的全局最优解。首次尝试将最小二乘支持向量机技术用于土壤侵蚀预测,并与BP神经网络的方法进行了对比,取得了较好的预测精度。

Research of small watershed water erosion prediction model based on SVM

Soil erosion is a very complicated process, and it is very difficult to predict with the equation. As a new machine learning algorithm, support vector translate the learning problem into a prediction quadratic problem to get a global optimization result with limited samples and the rule of minimize structure risk. This paper try to predict soil erosion with the Least Square support vector machine technology and the better predict precision compared to the BP artificial neural network has been gotten