Examination of the influence of different variables on prediction of unit cell parameters in perovskites using counter‐propagation artificial neural networks

Abstract

In this work, the unit cell parameter (a) of the series of cubic ABX 3 perovskites was modeled using counter-propagation artificial neural networks, and the influence of different input variables was examined by using algorithm for automatic adjustment of the relative importance of the variables. The input variables used in this model were the ionic radii of A, B, and X as well as the oxidation state (z) and the electronegativity (x) of the anion.The developed models have good generalization performances-good agreement between experimental and predicted values for lattice parameter. One of the important outcomes from this work is obtained from the results of the automatic adjustment of the relative importance of input variables. That is to say, this analysis gave us an insight that the most pronounced influence on the successful prediction of the unit cell parameter of the analyzed data set of cubic ABX 3 perovskites has the effective ionic radii of B-cation. In addition to this, it may be concluded that the separation of the compounds in different regions of counter-propagation artificial neural networks was predominantly influenced by the input variables with regard to the physical parameters of the anion. Copyright

Classification of urban soils for forensic purposes using supervised self‐organizing maps

“…GAs have been proven as an effective optimization tool 46–48 allowing relatively fast convergence without need of running every permutation of variables. In chemistry literature, the theory and use of GAs as a variable selection tool has been reported several times, 49–55 but also it has been used for automated optimization of different parameters of the models 39–45 . Here, only a brief summary of the main ideas of the GA algorithm as well as the procedure used in this work is explained.…”

“…In chemistry literature, the theory and use of GAs as a variable selection tool has been reported several times, [49][50][51][52][53][54][55] but also it has been used for automated optimization of different parameters of the models. [39][40][41][42][43][44][45] Here, only a brief summary of the main ideas of the GA algorithm as well as the procedure used in this work is explained.…”

“…This artificial neural network algorithm has previously been used for the analysis of IR spectra 37,38 . This algorithm together with some of its variants has been previously used in our laboratory 39–45 . To the best of our knowledge, SOMs have not been used previously for forensic purposes.…”

“…37,38 This algorithm together with some of its variants has been previously used in our laboratory. [39][40][41][42][43][44][45] To the best of our knowledge, SOMs have not been used previously for forensic purposes. The optimization of the parameters of the SOM as well as for the selection of the parts of the IR spectra that are most suitable for classification purposes, in this study, were performed in automated manner using genetic algorithm (GA).…”

Deep Learning Based Prediction of Perovskite Lattice Parameters from Hirshfeld Surface Fingerprints

“…This arises from the fact that the lattice parameter serves as a probe into the complex geometric and bonding characteristics that govern the stability and properties of such compounds. The studies have mainly involved the statistical/machine learning-based analysis of experimental data, augmented by additional descriptor data such as atomic radii, Pauling electronegativities, and valence or oxidation states. − From a machine learning perspective, these studies utilized traditional algorithms such as Support Vector Regression (SVR), Artificial Neural Network (ANN), and General Regression Neural Network (GRNN). These have achieved high accuracy for experimentally realized cubic perovskites but are rarely tested against the wide chemical space that Density Functional Theory (DFT) can access.…”