1997
DOI: 10.1016/s0097-8485(97)00019-3
|View full text |Cite
|
Sign up to set email alerts
|

13C NMR chemical shift sum prediction for alkanes using neural networks

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
2

Citation Types

0
11
0

Year Published

1999
1999
2018
2018

Publication Types

Select...
6
2

Relationship

0
8

Authors

Journals

citations
Cited by 18 publications
(11 citation statements)
references
References 26 publications
0
11
0
Order By: Relevance
“…However, for MLR equations, this technique induces a too small modification of the initial data set, and for the present study, we have estimated the stability and predictive ability of the MLR model with the leave-20%-out (L20%O) cross-validation method. [47][48][49] The L20%O cross-validation was applied by establishing a prediction set consisting of 20% of the patterns randomly selected from the entire set of experimental data and retaining the remaining 80% of the patterns in a calibration set. The MLR model is developed with the calibration set, and in a second phase, its coefficients are used to compute the viscosity of the molecules from the prediction set.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…However, for MLR equations, this technique induces a too small modification of the initial data set, and for the present study, we have estimated the stability and predictive ability of the MLR model with the leave-20%-out (L20%O) cross-validation method. [47][48][49] The L20%O cross-validation was applied by establishing a prediction set consisting of 20% of the patterns randomly selected from the entire set of experimental data and retaining the remaining 80% of the patterns in a calibration set. The MLR model is developed with the calibration set, and in a second phase, its coefficients are used to compute the viscosity of the molecules from the prediction set.…”
Section: Methodsmentioning
confidence: 99%
“…Because CODESSA does not offer this model validation option, the L20%O cross-validation was performed by exporting the data from CODESSA and doing the computations with the programs used in other QSPR investigations. [47][48][49] RESULTS AND DISCUSSION Using the HyperChem structural files and AMPAC AM1 output files for 369 compounds, CODESSA generated a set of 579 descriptors. Table 2 presents the notation of the 28 descriptors involved in the correlations reported in this paper.…”
Section: Methodsmentioning
confidence: 99%
“…Later, it was used to study compounds containing heteroatoms and/or rings. Over the last four decades, much effort has been focused on the development of 13 C NMR simulation, specifically for the following aspects: first, the development of computer science and computational chemistry have made 13 C NMR simulation to be an active field [6][7][8][9][10][11]; secondly, ab initio and density functional calculations for chemical shifts of organic molecules have recently emerged as one of the most promising new approaches for structure elucidation [12][13][14][15][16][17][18][19][20]; moreover, many computer systems have been developed for the prediction of NMR chemical shifts and for the elucidation of molecular structures. The combination of NMR chemical shift prediction systems and structural elucidation systems is becoming a powerful tool for automatic structural determination and identification [21,22]; in addition, both graph theory and topological theory are very powerful tools in NMR spectroscopic studies [23,24].…”
Section: Introductionmentioning
confidence: 99%
“…And recently, many mathematical, chemometric, statistic methods predicting 13 C NMR chemical shifts (CS) of organic compounds have been developed by means of artificial neural network [6][7][8][9][10][11][12][13][14][15] algorithm and/or multiple linear regression method [16]. Neural networks were also used to predict a group of alkanes [17][18][19]. Thereinto a lot of attention has been paid on employment of more general structural parameters, in particular, those derived from chemical graph theory [20][21][22][23][24][25][26][27].…”
Section: Introductionmentioning
confidence: 99%