【摘要】 建立了基于启发式方法和支持向量机方法的定量结构性质关系(QSPR)模型,用于预测α-环糊精与单取代或1,4-二取代苯衍生物结合后包结物的稳定常数．通过计算得到6个参数:分子重量、β-极化度、相对阳性电荷、相对阳性电荷表面积、DPSA3和分子轨道最大成键贡献,用于启发式方法和支持向量机方法建立QSPR模型,其相关系数分别是0．94和0．98,LOO交互检验的相关系数分别为0．92和0．95．因此,用支持向量机方法建立的模型要优于启发式方法,其预测能力更强、模型的稳定性更好．更多还原

【Abstract】 Support vector machine, as a novel machine learning technique, was used to construct QSPR model to describe the complexation ofα-cyclodextrin with mono-and 1, 4-disubstituted benzene derivative molecular descriptors. The association constants (Ka) for the inclusion complexation of cyclodextrins and benzene derivatives are calculated by the models found with a high degree of precision. The excellent prediction results with correlation coefficient of the heuristic method and support vector machines were 0.94 and 0.98 respectively. The cross-validation correlation coefficient of heuristic method and support vector machine were 0.92 and 0.95 respectively. We also found that six parameters of molecular weight, max bonding contribution of a MO, RPCG, RPCS, DPSA-3 and BETA polarizability can be used not only to predict Ka of the inclusion complexation of cyclodextrins and benzene derivatives but also to explain the mechanism of cyclodextrin combined with the guest. The advantages and disadvantages of two approaches were discussed, and it is concluded that the support vector machine is a better method to make QSPR models for predicting Ka.更多还原