检索规则说明:AND代表“并且”;OR代表“或者”;NOT代表“不包含”;(注意必须大写,运算符两边需空一格)
检 索 范 例 :范例一: (K=图书馆学 OR K=情报学) AND A=范并思 范例二:J=计算机应用与软件 AND (U=C++ OR U=Basic) NOT M=Visual
机构地区:[1]清华大学环境科学与工程系POPs研究中心,北京100084
出 处:《计算机与应用化学》2009年第6期773-777,共5页Computers and Applied Chemistry
基 金:"十一五"国家科技支撑计划课题"执行持久性有机污染物公约的支撑技术研究与示范"(课题编号:2007BAC03A09)
摘 要:应用ChemOffice 8.0中的MOPAC-PM3算法计算得到多溴联苯醚(PBDEs)的6个量子化学参数,采用基于多项式核,径向基核及Sigmoid核的支持向量机(SVM)方法建立了23种PBDEs蒸汽压的QSPR模型。三类核函数对训练集拟合的相关系数R^2分别为0.994,0.996,0.994,均方误差MSE分别为0.0102,0.0081,0.0095;留一法交叉验证(LOO)的相关系数分别为0.992,0.991,0.991.对测试集进行回归的相关系数分别为0.994,0.986,0.991,均方误差MSE分别为0.0225,0.0458,0.0247。结果表明SVM回归算法在PBDEs蒸汽压的QSPR建模上表现出色,核函数的选择对SVM模型性能影响不显著。文章还从模型拟合效果、预测能力及稳定性三方面比较了三类核函数的整体性能,并比较了支持向量数目对核函数预测能力的影响。结果表明多项式核与Sigmoid核性能相当,优于径向基核。With MOPAC-PM3 method in the ChemOffice 8.0 software, 6 quantum chemistry parameters for 23 kinds of PBDEs were calculated. Using support vector machine (SVM) method based on polynomial kernel, radial basis kernel, sigmoid kernel, a QSPR model has been set up for vapor pressure of 23 kinds of PBDEs. For the training set, the correlation coefficient R2 of the three types of kernel function is 0. 994, 0. 996 and 0. 994, the mean square error (MSE) is 0. 0102, 0. 0081 and 0. 0095, respectively. The correlation coefficient R2 of Leave One Out (LOO) cross validation is 0. 992, 0. 991 and 0. 991, respectively. For the testing set, the correlation coefficient R~ of the three types of kernel function is 0. 994, 0. 986 and 0. 991, the MSE is 0. 0225, 0. 0458 and 0. 0247, respectively. The results show that the SVM regression algorithm play an excellent performance on QSPR modeling for the vapor pressure of PBDEs, and the choice of kernel function of SVM model do not significantly affect performance. The overall performance of the three types of kernel functions has been compared through model fitting, prediction ability and the stability, and compared the influence of the support vector number on the prediction ability. The results show that the properties of polynomial kernel and sigmoid kernel are similar, better than radial basis kernel.
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在载入数据...
正在链接到云南高校图书馆文献保障联盟下载...
云南高校图书馆联盟文献共享服务平台 版权所有©
您的IP:3.17.164.48