论文部分内容阅读
支持向量机(SVM)算法是特别适合于用有限已知样本训练建模,进而预报未知样本属性的模式识别新算法。本文旨在尝试将Vapnik提出的支持向量机算法用于环境保护领域。用支持向量回归算法总结了石城岛、王家岛附近赤潮发生与海水温度、溶解氧、盐度、总氮量、无机磷、浮游植物密度的对应关系。用支持向量回归算法求得赤潮爆发的数学模型。留一法结果表明:支持向量回归的预报误差比人工神经网络小。支持向量机方法可以成为研究赤潮发生机理、探索赤潮预报途径的一种工具。
Support Vector Machine (SVM) algorithm is a new pattern recognition algorithm which is especially suitable for training modeling with limited known samples and then predicting the unknown sample attributes. This paper attempts to use Vapnik’s SVM algorithm in the field of environmental protection. The relationship between the occurrence of red tide and the seawater temperature, dissolved oxygen, salinity, total nitrogen, inorganic phosphorus and phytoplankton density was summarized with the support vector regression algorithm. Using Support Vector Regression Algorithm to Obtain Mathematical Model of Red Tide Explosion. The result of leaving one method shows that the prediction error of support vector regression is smaller than that of artificial neural network. SVM can be a tool to study the mechanism of red tide and explore the way of forecasting red tide.