Abstract: The structure parameters of the quantum chemistry forpolychlorinated dibenzofurans (PCD-Fs) compounds were calculated by using theMOPAC-AM1 method in Chemoffice 8.0.Some parameters are selected as the structuredescriptors of PCDFs compounds.The molecular structures and the model of n-octanol/water partition coefficients are constructed and predicted in terms of back-propagation network and radial basis function networks in artificial neuralnetwork.These results are compared with the results of multiple regression methods.It can be found that the results of back-propagation network and radial basis function networks are better than those of multiple regression methods.

Key words: artificial neural network, polychlorinated dibenzofurans, n-octanol/water partition coefficients