Prediction of physicochemical properties from Euclidean distance methods based on electrotopological state indices

This paper describes predictions of log K-OW, Henry's Law constant, vapor pressure, and OH-radical bimolecular rate constant from two Euclidean distance methods, using electrotopological state indices as input. The quality of the predictions is highly dependent on the size of the experimental dataset, i.e., the density of experimental data points in the Euclidean space. The Euclidean distance results are compared with predictions made by commercial software products for the same sets of compounds. (C) 1999 Elsevier Science Ltd. All rights reserved.