Genetic Algorithm-based Multivariate Linear Regression Research Articles

A molecular-based model for prediction of liquid viscosity of pure organic compounds: A quantitative structure property relationship (QSPR) approach

In this study, a new quantitative structure–property relationship (QSPR) is presented for prediction of the liquid viscosity of pure organic compounds. The model implements eight molecular descriptors selected using the genetic algorithm-based multivariate linear regression (GA-MLR) from more than various 3000 molecular descriptors to predict the liquid viscosity. To propose a comprehensive and predictive model, 2748 pure organic compounds are investigated. Furthermore, several statistical methods are applied to evaluate the predictive capability of the model. The root mean square of error and the average absolute percent error of the model are equal to 0.34 and 7%, respectively.

Determination of normal boiling vaporization enthalpy using a new molecular-based model

In this study, a new eight-parameter equation is presented for calculation/prediction of the normal boiling vaporization enthalpy of pure chemical compounds. The model variables include seven molecular-based descriptors in addition to the normal boiling point temperature. The model is developed using the Genetic-Algorithm-based Multivariate Linear Regression (GA-MLR) method. The GA-MLR is applied to select the parameters of the model from more than 3000 molecular-based parameters in addition to the normal boiling point. To propose a comprehensive and predictive model, 4879 pure chemical compounds are investigated. Furthermore, several statistical methods are implemented to evaluate the predictive power of the model. The root mean square of error and the average absolute deviation of the model are approximately equal to 1.8kJ/mol and 2.3%.

Phase Equilibrium Modeling of Structure H Clathrate Hydrates of Methane + Water “Insoluble” Hydrocarbon Promoter Using QSPR Molecular Approach

In this communication, the quantitative structure–property relationship (QSPR) strategy is applied to present two molecular models for determination of the structure H (sH) hydrate dissociation conditions with methane as help gas. Twenty-one water “insoluble” hydrocarbon promoters are examined. To propose reliable models, almost all of the available literature data are studied. Two mathematical methods including the genetic-algorithm-based multivariate linear regression (GA-MLR) and the least square support vector machines (LSSVM) are applied for determination (selection) of the model parameters. As a result, two reliable models are developed: (1) QSPR-GA-MLR linear model and (2) QSPR-LSSVM nonlinear model with satisfactory results quantified by the following statistical parameters: absolute average deviations (AAD) of the represented/predicted hydrate dissociation pressures from existing experimental values: about 9%, and squared correlation coefficient: 0.956 in the case of using the first model, and about 4% and 0.992 through applying the second model, respectively. These results demonstrate much better accuracy through the QSPR-LSSVM nonlinear model than applying the QSPR-MLR linear one.

A quantitative structure–property relationship for determination of enthalpy of fusion of pure compounds

In this study, the quantitative structure–property relationship method is applied to predict the enthalpy of fusion of pure chemical compounds at their normal melting point. A genetic algorithm-based multivariate linear regression is used to select the most statistically effective molecular descriptors for evaluating this property. To propose a comprehensive and predictive model, 3,846 pure chemical compounds are investigated. The root mean square of error and the average absolute deviation of the model are equal to 2.57 kJ/mol and 9.7%.

QSPR approach for determination of parachor of non-electrolyte organic compounds

In this work, the Quantitative Structure–Property Relationship (QSPR) method is applied to represent/predict the parachor of pure non-electrolyte organic compounds. A Genetic-Algorithm-based Multivariate Linear Regression (GA-MLR) is used to select the most statistically effective molecular descriptors for evaluating this property. To propose a predictive model, 227 pure non-electrolyte organic compounds are investigated. 2.8% absolute average deviation for the represented/predicted parachors of investigated compounds is found from the corresponding experimental values.

Prediction of Triple-Point Temperature of Pure Components Using their Chemical Structures

A quantitative structure property relationship study was performed to develop a model for the prediction of triple-point temperature of pure components. For developing this model, 638 pure components were used, and, for whichever, 1664 molecular descriptors were determined. As a standard tool for subset variable selection, genetic algorithm-based multivariate linear regression (GA-MLR) technique was used. The obtained model is a seven parameters multilinear equation that has a squared correlation coefficient of 0.9410 (R2 = 0.9410).

Estimation of molecular diffusivity of pure chemicals in water: a quantitative structure–property relationship study

A quantitative structure–property relationship (QSPR) study was performed to predict the molecular diffusivity of pure chemicals in water. A genetic-algorithm-based multivariate linear regression (GA-MLR) was applied to select the most statistically effective molecular descriptors for modelling the molecular diffusivity of pure chemicals in water. Based on the selected molecular descriptors, a three-layer feed forward neural network (FFNN) was constructed to predict the property. The obtained results showed that the FFNN was able to predict the molecular diffusivity of pure chemicals in water.

Estimation of Aniline Point Temperature of Pure Hydrocarbons: A Quantitative Structure−Property Relationship Approach

In the present work, a quantitative structure−property relationship (QSPR) study is performed to predict the aniline point temperature of pure hydrocarbon components. As a powerful tool, genetic algorithm-based multivariate linear regression (GA-MLR) is applied to select most statistically effective molecular descriptors on the aniline point temperature of pure hydrocarbon components. Also, a three-layer feed forward neural network (FFNN) is constructed to consider the nonlinear behavior of appearing molecular descriptors in GA-MLR result. The obtained results show that the constructed FFNN can accurately predict the aniline point temperature of pure hydrocarbon components.

Prediction of some important physical properties of sulfur compounds using quantitative structure–properties relationships

In this work, physical properties of sulfur compounds (critical temperature (Tc), critical pressure (Pc), and Pitzer's acentric factor (omega)) are predicted using quantitative structure-property relationship technique. Sulfur compounds present in petroleum cuts are considered environmental hazards. Genetic algorithm based multivariate linear regression (GA-MLR) is used to select most statistically effective molecular descriptors on the properties. Using the selected molecular descriptors, feed forward neural networks (FFNNs) are applied to develop some molecular-based models to predict the properties. The presented models are quite accurate and can be used to predict the properties of sulfur compounds.

Prediction of molecular diffusivity of pure components into air: A QSPR approach

The molecular diffusivity of 378 pure components into air was predicted using genetic algorithm-based multivariate linear regression (GA-MLR) and feed forward neural networks (FFNN). GA-MLR was used to select the molecular descriptors, as inputs for FFNN. The correlation coefficient ( R 2 ) of obtained multivariate linear seven-descriptor model by GA-MLR is 0.9334 and the same value for generated FFNN is 0.9643. These models can be applied for prediction of molecular diffusivity of pollutants into air in case of air pollution studies.

Prediction of the Watson Characterization Factor of Hydrocarbon Components from Molecular Properties

AbstractIn the present work, a Quantitative Structure–Property Relationship (QSPR) study was performed to predict the Watson characterization factor of hydrocarbon components. A Genetic Algorithm‐based Multivariate Linear Regression (GA‐MLR) was applied to select the most statistically effective molecular descriptors of the Watson characterization factor. Then, based on the selected molecular descriptors by GA‐MLR, a three‐layer Feed Forward Neural Network (FFNN) was constructed to predict the Watson characterization factor. The obtained results showed that the constructed FFNN can accurately predict the Watson characterization factor of hydrocarbon components.

Prediction of Flash Point Temperature of Pure Components Using a Quantitative Structure–Property Relationship Model

AbstractIn this work, a general Quantitative Structure–Property Relationship (QSPR) model ($\rm{ R^2 = 0.9669}$, $\rm{ Q_{{\rm{LOO}}}^{\rm{2}} = 0.9663}$, and $\rm{ s = 12.691}$) for the prediction of flash points of 1030 pure compounds is developed. Genetic Algorithm‐based Multivariate Linear Regression (GA‐MLR) technique is used to select four chemical structure‐based molecular descriptors from a pool containing 1664 molecular descriptors.

QSPR Studies for Solubility Parameter by Means of Genetic Algorithm‐Based Multivariate Linear Regression and Generalized Regression Neural Network

AbstractThe solubility parameters of 1228 solvents, from all the chemical groups, were predicted using Genetic Algorithm‐Based Multivariate Linear Regression (GA‐MLR) and Generalized Function Approximation Neural Network (GRNN). GA‐MLR was used to select the molecular descriptors, as inputs for GRNN. The obtained multivariate linear seven descriptors model by GA‐MLR had a correlation coefficient of $\rm{ R^2 = 0.821}$. The generated GRNN in this work has a correlation coefficient of $\rm{ R^2 = 0.98}$.

A Molecular‐Based Model for Prediction of Solubility of C60 Fullerene in Various Solvents

In this presented work, a quantitative structure‐property relationship study (QSPR) was done for prediction of solubility of C60 fullerene in various solvents. In this study, genetic algorithm‐based multivariate linear regression (GA‐MLR) was applied to obtain most statistically effective molecular descriptors on solubility of C60 in various solvents. All of these molecular descriptors are only calculated from the chemical structure of solvents. For considering nonlinear behavior of appearing molecular descriptors in GA‐MLR section, a feed forward neural network (FFNN) was constructed and optimized for prediction of solubility of C60 fullerene in solvents. Obtained models considerably showed better accuracy in comparison with the previous models.

A new molecular-based model for prediction of enthalpy of sublimation of pure components

A quantitative structure property relationship (QSPR) study was performed to develop a model for prediction of enthalpy of sublimation of pure components. For developing this model, 1348 pure components were used, and for each of them, 1664 molecular descriptors were determined. As a standard tool for subset variable selection, genetic algorithm-based multivariate linear regression (GA-MLR) technique was used. The obtained model is a five-parameter multi-linear equation that has a squared correlation coefficient of 0.9746 ( R 2 = 0.9746).

A New Accurate Neural Network Quantitative Structure- Property Relationship for Prediction of θ (Lower Critical Solution Temperature) of Polymer Solutions

Abstract In this study, a new neural network quantitative structure-property relationship model for prediction of θ (LCST ) of polymer solutions is presented. The parameters of this model are eight molecular descriptors which are calculated only from the chemical structure of polymer and solvent. These eight molecular descriptors were selected from 3328 molecular descriptors of polymer and solvent available in polymer solution by genetic algorithm-based multivariate linear regression (GA-MLR) technique. The obtained neural network model can predict the θ (LCST ) of 169 polymer solutions with mean relative error of 1.67% and squared correlation coefficient of 0.9736.

Prediction of Standard Enthalpy of Formation by a QSPR Model

The standard enthalpy of formation of 1115 compounds from all chemicalgroups, were predicted using genetic algorithm-based multivariate linear regression (GA-MLR). The obtained multivariate linear five descriptors model by GA-MLR has correlationcoefficient ( R 2 = 0.9830 ). All molecular descriptors which have entered in this model arecalculated from chemical structure of any molecule. As a result, application of this modelfor any compound is easy and accurate.

QSPR analysis for intrinsic viscosity of polymer solutions by means of GA-MLR and RBFNN

A quantitative structure–property relationship (QSPR) treatment of intrinsic viscosity of polymer solutions was performed by means of a genetic algorithm based multivariate linear regression (GA-MLR). A five parameters correlation, with squared correlation coefficient R2=0.8275 gives good predictions for 65 polymer solutions. In preparation of this model, 1664 molecular descriptors for each polymer and 1664 molecular descriptors for each solvent were checked and finally, five molecular descriptors were selected. For considering the nonlinear behavior of these five molecular descriptors, a radial based function neural network (RBFNN) with squared correlation coefficient R2=0.9100 was constructed. Notably, all the parameters involved in these equations can be derived solely from the chemical structure of the polymers repeating unit and the solvents which makes them very useful for prediction of the intrinsic viscosity of unknown or unavailable polymer solutions.