Different sets of molecular descriptors using the k-nearest neighbor classification method were used to make a general classification of 152 organic solvents, the classification being further improved by performing the counter-propagation artificial neural network. An extensive investigation was made of the physico-chemical properties of 152 solvents in a search for quantitative structure-property relationships (QSPR). Wide sets of molecular descriptors were tested and regression models were obtained by selecting the best descriptor subset by genetic algorithm in order to optimize their prediction power
Gramatica, P., Navas, N., Todeschini, R. (1999). Classification of organic solvents and modelling of their physico-chemical properties by chemometric methods using different sets of molecular descriptors. TRAC. TRENDS IN ANALYTICAL CHEMISTRY, 18(7), 461-471 [10.1016/S0165-9936(99)00115-6].
Classification of organic solvents and modelling of their physico-chemical properties by chemometric methods using different sets of molecular descriptors
TODESCHINI, ROBERTO
1999
Abstract
Different sets of molecular descriptors using the k-nearest neighbor classification method were used to make a general classification of 152 organic solvents, the classification being further improved by performing the counter-propagation artificial neural network. An extensive investigation was made of the physico-chemical properties of 152 solvents in a search for quantitative structure-property relationships (QSPR). Wide sets of molecular descriptors were tested and regression models were obtained by selecting the best descriptor subset by genetic algorithm in order to optimize their prediction powerI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.