NEURAL NETWORK CLASSIFICATION OF INDUCTIVE AND RESONANCE EFFECTS OF SUBSTITUENTS

被引:16
作者
KVASNICKA, V
SKLENAK, S
POSPICHAL, J
机构
[1] Department of Mathematics, Faculty of Chemical Technology, Slovak Technical University
关键词
D O I
10.1021/ja00057a039
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
An application of feed-forward neural networks adapted by back-propagation strategy to classification of inductive and resonance effects of functional groups is described. The descriptors of functional groups, forming the input information to the neural network, are determined as numbers of times certain subgraphs (with evaluated vertices and edges) appear in the functional group under study. The bidden activities are used for cluster analysis of functional groups.
引用
收藏
页码:1495 / 1500
页数:6
相关论文
共 31 条