Abductive networks: generalization, pattern recognition, and prediction of chemical behavior

Using commercially available software, it is possible to reduce numerical data to a mathematical representation called an abductive network (AN). In the current communication, we describe several simple examples which illustrate the interesting, and potentially useful properties of abductive networks. We show that when applied to the correlation of Kovats indices with molecular refractivities and dipole moments of substituted phenols, abductive networks more accurately predict Kovats indices than do counter-propagation neural networks or linear regression equations. When applied to the modeling of quantitative structure–activity relationships (QSAR) for local anesthetics, AN's are marginally superior to regression. AN's offer the advantage that correlations may be drawn between variables which are not easily related within a mathematical context.