MULTIPLE LINEAR REGRESSION-ARTIFICIAL NEURON NETWORK MIXED MODEL FOR PREDICTING THE LIQUID DENSITY AT NORMAL BOILING POINT OF A PURE ORGANIC COMPOUND CAPABLE OF FORMING AN ANN OUTPUTTING THE STANDARD STATE ENTHALPY OF FORMATION BY USING A MOLECULE DESCRIPTOR INCLUDED IN AN MLRM

摘要

PURPOSE: An MLR(Multiple Linear Regression)-ANN(Artificial Neuron Network) mixed model for predicting the liquid density at normal boiling point of a pure organic compound is provided to form an ANN outputting the liquid density at normal boiling point by using a molecule descriptor included in an MLRM(Multiple Linear Regression Model), thereby improving prediction performance.;CONSTITUTION: A molecule descriptor value about the liquid density at normal boiling point of sample organic compounds is prepared. Experimental data is separated into a training set and a test set. An optimum MLRM(Multiple Linear Regression Model) for the training set is explored. After an optimum ANNM(Artificial Neural Network Model) divides every samples into three sets, it is explored. If the absolute value of the difference of a liquid density prediction value, figured out by the MLRM and the ANNM, is greater than an over-suitability preventing standard value, the liquid density prediction value by the MLRM is selected as a liquid density value.;COPYRIGHT KIPO 2012