Neural networks studies : quantitative structure-activity relationship of mutagenic aromatic nitro compounds
Didier Villemin, Driss Cherqaoui, JM. Cense · Journal de Chimie Physique · 1993
The application of neural networks to the study of quantitative structure-activity relationship (QSAR) of mutagenic aromatic and heteroaromatic nitro compounds is reported. The results obtained are compared with the results given by a multiple linear regression. It is shown that neural networks prediction is more accurate than regression analysis prediction.