APPLICATION OF NEURAL NETWORKS TO STRUCTURE SANDALWOOD ODOR RELATIONSHIPS

Neural networks have proved to be particularly successful in their abi lity to identify non-linear relationships. This paper shows that a thr ee-layer back-propagation neural network is able to learn the relation ship between the sandalwood odour and molecular structures of 85 organ ic compounds belonging to acyclic, cyclohexyl, norbornyl, campholenyl and dec alin derivatives. Four steric and three electronic parameters were used to describe each molecular structure, Odour was coded by a b inary variable, The neural network was used to classify the compounds into two groups and to predict their odours (sandalwood or non-sandalw ood). The results obtained were compared with those given by discrimin ant analysis, and found to be better, The most important descriptors w ere revealed on the basis of correlation analysis. (C) 1997 by John Wi ley & Sons, Ltd.