支持向量机算法用于夜光藻密度建模

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 applied to modelling of density of noctiluca scientillans in sea water

Support vector machine proposed by Vapnik is a newly developed technique for data mining. It is suitable for the datu processing based on finite number of training samples, with special technique to restrict overfitting. In this work, support vector classification and regres-sion techniques are used to make modeling on the relationship between the density of noctiluca scientillans and the temperature, salt concentra-tion, dissolved oxygen and the content of nitrogen, phorsphorus and plant organism in sea water. It has been found that SVC and SVR can give better modeling results than that of ANN and traditional linear regression.