A novel method for examination of the variable contribution to computational neural network models

Computational neural networks (CNNs or, as they are commonly referred to; a rtificial neural networks, ANNs) have been demonstrated in a large number o f applications to be useful for modeling and prediction. They suffer, howev er, in their conventional use, that is feed forward/back-propagation of the error, from the lack of a simple or straightforward means of interpreting the variable contribution to the models. CNNs are therefore often referred to as black box models. In this study novel algorithmic approaches to the i nterpretation of CNN models are proposed, examined and compared with the co rresponding variable contribution in partial least squares (PLS) regression models. A sensitive analysis of the CNN models is carried out by sequentia lly setting each input variable to zero. In addition, to evaluate the direc tion of the variable contribution, the linear regression coefficients for e ach input variable are generated. The results of these two approaches are t hen combined to facilitate comparison with PLS models. CNN models for data on chiral separation, 3D-QSRR (quantitative structure-retention relationshi ps) and SIMS (secondary ion mass spectroscopy) are used to demonstrate the feasibility of the method. For the latter two data sets, there is close agr eement between the PLS and CNN models with regard to variable contribution. For the nonlinear data set for chiral separation, differences in variable contribution are revealed. (C) 1998 Elsevier Science B.V. All rights reserv ed.