Neural network for prediction of ¹³C NMR chemical shifts of fullerene C₆₀ mono‐adducts

Abstract

Real‐valued models based on deep artificial neural networks were proposed to predict 13C NMR chemical shifts of fullerene C60 core carbon atoms for computer‐aided structure elucidation of complex fullerene C60 mono‐adducts. We showed that parametric rectified linear units could be successfully used as activation functions in hidden layers of artificial neural networks for decision of complex physical‐chemical tasks. A total of 400 artificial neural networks were trained and tested in order to reveal the best‐fitted models. The best prediction accuracy of real‐valued models was achieved with MAEP = 1.83 ppm/RMSEP = 2.60 ppm using artificial neural network model which has 110 and 120 hidden units, respectively, with parametric rectified linear unit as activation function.

Data Science Applied to Carbon Materials: Synthesis, Characterization, and Applications

“…In this sense, an ML model has been created for the analysis of nuclear magnetic resonance to assign chemical shifts of new fullerene derivatives by combining data from literature and complemented with new simulated data via density functional theory. [ 50 ]…”

Toward Accurate Predictions of Atomic Properties via Quantum Mechanics Descriptors Augmented Graph Convolutional Neural Network: Application of This Novel Approach in NMR Chemical Shifts Predictions

“…Graph Convolutional Network (GCN) is a promising tool for computational chemists to conduct molecular properties predictions. , The advantage of this neural network lies in the fact that it could efficiently extract the useful information from molecular structures, without applying complex descriptors. However, to successfully transfer such a novel tool for atomic properties predictions, in one aspect, the neural network’s architecture should be designed with respect to practical purpose, , and in another aspect, some related atomic descriptors should be included to refine the model. Both of these are hurdles for computational chemists.…”