简介:在这个工作,二个chemometrics方法被申请建模和一些器官、无机的混合物的electrophoretic活动性的预言。连续设计算法,特征选择(矿泉)策略,被用作描述符选择和模型发展方法。然后,支持向量机器(SVM)和多重线性回归(MLR)当模特儿被利用构造非线性、线性的量的structureproperty关系模型。用SVM模型获得的结果与用MLR获得的那些相比表明SVM模型比MLR那个具有更好预兆的价值。分别地,当时,为训练集合的root-mean-square错误和为SVM模型的测试集合是0.1911和0.2569由MLR模型,他们分别地是0.4908和0.6494。结果证明SVM模型急速地在QSPR研究提高预言的能力并且比MLR模型优异。
简介:Inthiswork,somechemometricsmethodsareappliedforthemodelingandpredictionoftheHildebrandsolubilityparameterofsomepolymers.Ageneticalgorithm(GA)methodisdesignedfortheselectionofvariablestoconstructtwomodelsusingthemultiplelinearregression(MLR)andleastsquare-supportvectormachine(LS-SVM)methodsinordertopredicttheHildebrandsolubilityparameter.TheMLRmethodisusedtobuildalinearrelationshipbetweenthemoleculardescriptorsandtheHildebrandsolubilityparameterforthesecompounds.ThentheLS-SVMmethodisutilizedtoconstructthenon-linearquantitativestructure-activityrelationship(QSAR)models.TheresultsobtainedusingtheLS-SVMmethodarethencomparedwiththoseobtainedfortheMLRmethod;itwasrevealedthattheLS-SVMmodelwasmuchbetterthantheMLRone.Theroot-mean-squareerrorsofthetrainingsetandthetestsetfortheLS-SVMmodelwere0.2912and0.2427,andthecorrelationcoefficientswere0.9662and0.9518,respectively.ThispaperprovidesanewandeffectivemethodforpredictingtheHildebrandsolubilityparameterforsomepolymers,andalsorevealsthattheLS-SVMmethodcanbeusedasapowerfulchemometricstoolforthequantitativestructure-propertyrelationship(QSPR)studies.