2013
DOI: 10.1080/00319104.2012.708932
|View full text |Cite
|
Sign up to set email alerts
|

Application of QSPR for the prediction of gas to 1-octanol solvation enthalpy using support vector regression

Abstract: A quantitative structure property relationship model was developed to predict gas to 1-octanol solvation enthalpy (DH Solv ) of 127 different organic compounds using support vector machine (SVM). The variable selection method of genetic algorithm (GA) was employed to select optimal subset of descriptors. The five descriptors selected by GA were used as inputs for construction of the multiple linear regression (MLR), artificial neural network (ANN) and SVM models. The standard errors of for the prediction data … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

1
4
0

Year Published

2014
2014
2023
2023

Publication Types

Select...
6

Relationship

2
4

Authors

Journals

citations
Cited by 9 publications
(5 citation statements)
references
References 46 publications
1
4
0
Order By: Relevance
“…The reported correlation coefficients for these models [31,54] are similar to, or greater than what we report for Model 2 and the standard errors of the models developed using a support vector machine method and artificial neural networks is smaller than the RMSE of Model 2. However, these models cannot be used easily in high throughput applications because acquiring the molecular descriptors requires multiple software applications [54], which can be time-consuming and lends itself to increased chances of human error.…”
Section: Comparison With Other 1u • Oa Modelssupporting
confidence: 75%
See 2 more Smart Citations
“…The reported correlation coefficients for these models [31,54] are similar to, or greater than what we report for Model 2 and the standard errors of the models developed using a support vector machine method and artificial neural networks is smaller than the RMSE of Model 2. However, these models cannot be used easily in high throughput applications because acquiring the molecular descriptors requires multiple software applications [54], which can be time-consuming and lends itself to increased chances of human error.…”
Section: Comparison With Other 1u • Oa Modelssupporting
confidence: 75%
“…If we compare the performance of the Mintz et al [19] model and Model 2 when using all 195 chemicals in the ΔU • OA dataset, we see that Model 2 performs better than the Mintz et al [19] model (Table 4). While there is little difference between the statistics on the residuals, fewer chemicals have residuals greater than 10 kJ•mol −1 when using Model In addition to the model by Mintz et al [19], there is a ppLFER for wet-octanol air ΔH • values [31] and models for ΔH • OA using a support vector machine method, artificial neural networks, and MLRs based on various molecular descriptors [54]. The wet-octanol air ΔH • OA model is based on the same dataset used to develop the Mintz et al [19] model above [31].…”
Section: Comparison With Other 1u • Oa Modelsmentioning
confidence: 99%
See 1 more Smart Citation
“…Support vector machine (SVM) is a novel kind of machine learning method, and is gaining reputation owing to many attractive features as well as talented empirical performance. SVMs have found numerous applications in prediction in QSPR studies [24][25][26][27][28][29][30].…”
Section: Introductionmentioning
confidence: 99%
“…Due to its wonderful generalization performance, the SVM has attracted interest and obtained extensive application [13][14][15][16][17][18][19] . In recent years, SVM has also exposed enormous performance in QSPR studies owing to its ability to construe the nonlinear relationships between molecular structure and properties [20][21][22][23][24][25] .…”
Section: Introductionmentioning
confidence: 99%