ISSN print edition: 0366-6352
ISSN electronic edition: 1336-9075
Registr. No.: MK SR 9/7
Artificial neural network prediction of steric hindrance parameter of polymers
Xinliang Yu, Wenhao Yu, Bing Yi, and Xueye Wang
College of Chemistry and Chemical Engineering, Hunan Institute of Engineering, Xiangtan, Hunan, 411104, China
Received: 5 October 2008 Revised: 16 November 2008 Accepted: 19 November 2008
Abstract: An artificial neural network (ANN) model for modeling and prediction of the steric hindrance parameter σ of polymers with
three quantum chemical descriptors, the average polarizability of a molecule α, entropy S, and dipole moment μ, was developed. These descriptors were calculated from the monomers of the respective polymers according
to the density functional theory at the B3LYP level of the theory with the 6-31G(d) basis set. Optimal conditions were obtained
by adjusting various parameters by trial-and-error. Simulated with the final optimum BP neural network [3-1-1], the results
show that the predicted σ values are in good agreement with the experimental ones, with the root mean square (rms) error being
0.080 (R = 0.945) for the training set, and 0.078 (R = 0.918) for the test set, which indicates that the proposed model has better predictive capability than the existing one.
Keywords: artificial neural network - DFT - QSPR - steric hindrance parameter
Full paper is available at www.springerlink.com.
Chemical Papers 63 (4) 432–437 (2009)