论文部分内容阅读
讨论了用人工神经网络法预测物性的几类输入参数．用分子描述码作为输入参数预测了烯烃的沸点，烯烃（链）的相关系数ｒ＝０．９９７５，标准差Ｓ＝１．６８７．环烯烃：相关系数ｒ＝０．９９７０，标准差Ｓ＝２．５４８．优于三维连接性指数的计算值．
Several types of input parameters for predicting physical properties using artificial neural networks are discussed. The molecular descriptors were used as input parameters to predict the boiling point of olefins. The correlation coefficient of olefin (chain) was 0.9975 and the standard deviation S was 1.687. Cyclic olefins: correlation coefficient r = 0.9970, standard deviation S = 2.548. Better than the calculated value of the three-dimensional connectivity index.