论文部分内容阅读
A molecular vector-type descriptor containing 6 variables is used to describe the structure of aromatic hydrocarbons (AHs) and relate to normal boiling points (bp) of AHs. The correlation coefficient (R) between the estimated bp and experimental bp is 0.9988 and the root mean square error (RMS) is 7.907(C for 66 AHs. The RMS obtained by cross-validation is 9.131(C, which implies the relationship model having good prediction ability.
A molecular vector-type descriptor containing 6 variables is used to describe the structure of aromatic hydrocarbons (AHs) and relate to normal boiling points (bp) of AHs. The correlation coefficient (R) between the estimated bp and experimental bp is 0.9988 and the root mean square error (RMS) is 7.907 (C for 66 AHs. The RMS obtained by cross-validation is 9.131 (C, which implies the relationship model having good prediction ability.