scholarly journals Crystal-Site-Based Artificial Neural Networks for Material Classification

Crystals ◽  
2021 ◽  
Vol 11 (9) ◽  
pp. 1039
Author(s):  
Juan I. Gómez-Peralta ◽  
Nidia G. García-Peña ◽  
Xim Bokhimi
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Crystal Structures ◽  
Materials Science ◽  
Layered Perovskite ◽  
Structure Property ◽  
Average Precision ◽  
Artificial Neural ◽  
Spinel Structures ◽  
Crystal Compound

In materials science, crystal structures are the cornerstone in the structure–property paradigm. The description of crystal compounds may be ascribed to the number of different atomic chemical environments, which are related to the Wyckoff sites. Hence, a set of features related to the different atomic environments in a crystal compound can be constructed as input data for artificial neural networks (ANNs). In this article, we show the performance of a series of ANNs developed using crystal-site-based features. These ANNs were developed to classify compounds into halite, garnet, fluorite, hexagonal perovskite, ilmenite, layered perovskite, -o-tp- perovskite, perovskite, and spinel structures. Using crystal-site-based features, the ANNs were able to classify the crystal compounds with a 93.72% average precision. Furthermore, the ANNs were able to retrieve missing compounds with one of these archetypical structure types from a database. Finally, we showed that the developed ANNs were also suitable for a multitask learning paradigm, since the extracted information in the hidden layers linearly correlated with lattice parameters of the crystal structures.

scholarly journals Application of Artificial Neural Networks to Predict the Intrinsic Solubility of Drug-Like Molecules

Pharmaceutics ◽  
2021 ◽  
Vol 13 (7) ◽  
pp. 1101
Author(s):  
Elena M. Tosca ◽  
Roberta Bartolucci ◽  
Paolo Magni
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Molecular Descriptor ◽  
Biological Properties ◽  
Training Dataset ◽  
Structure Property ◽  
Feed Forward Neural Network ◽  
Qspr Model ◽  
Artificial Neural ◽  
Intrinsic Solubility

Machine learning (ML) approaches are receiving increasing attention from pharmaceutical companies and regulatory agencies, given their ability to mine knowledge from available data. In drug discovery, for example, they are employed in quantitative structure–property relationship (QSPR) models to predict biological properties from the chemical structure of a drug molecule. In this paper, following the Second Solubility Challenge (SC-2), a QSPR model based on artificial neural networks (ANNs) was built to predict the intrinsic solubility (logS0) of the 100-compound low-variance tight set and the 32-compound high-variance loose set provided by SC-2 as test datasets. First, a training dataset of 270 drug-like molecules with logS0 value experimentally determined was gathered from the literature. Then, a standard three-layer feed-forward neural network was defined by using 10 ChemGPS physico-chemical descriptors as input features. The developed ANN showed adequate predictive performances on both of the SC-2 test datasets. Benefits and limitations of ML approaches have been highlighted and discussed, starting from this case-study. The main findings confirmed that ML approaches are an attractive and promising tool to predict logS0; however, many aspects, such as data quality, molecular descriptor computation and selection, and assessment of applicability domain, are crucial but often neglected, and should be carefully considered to improve predictions based on ML.

Neural network investigation of borate crystal properties

2002 ◽  
Vol 217 (5) ◽  
Author(s):  
M. J. Barber ◽  
P. Becker
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Networks ◽  
Crystal Structures ◽  
Chemical Properties ◽  
Crystal Chemical ◽  
Borate Crystal ◽  
Crystal Properties ◽  
Artificial Neural

AbstractCorrelations between crystal chemical properties of anhydrous oxoborate crystals wereanalysed using artificial neural networks. Using structuralproperties of oxoborate crystal structures described inthe literature, we developed several neural network modelsthat capture statistical relations between crystalchemical properties of the anhydrous oxoborates from the existingdata sets. This indicates the suitability of neural networks forthe prediction of structural propertiesof crystals.

Artificial Neural Networks in Materials Science Application

2010 ◽  
Vol 20-23 ◽  
pp. 1211-1216 ◽  
Author(s):  
Wen Yu Zhang
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Materials Science ◽  
Material Design ◽  
Widespread Application ◽  
Plastic Processing ◽  
Artificial Neural ◽  
Science Application ◽  
Compound Materials ◽  
Material Preparation

Because but the artificial neural networks has the strong non-linear problem handling ability also the fault tolerance strong obtains the widespread application in the materials science.This article to its material design, the material preparation craft optimizes, the plastic processing, the heat treatment, the compound materials, corrode, domain and so on casting applications have carried on the discussion.

scholarly journals Exploring Correlations Between Properties Using Artificial Neural Networks

2019 ◽  
Vol 51 (1) ◽  
pp. 58-75
Author(s):  
Yiming Zhang ◽  
Julian R. G. Evans ◽  
Shoufeng Yang
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Physical Properties ◽  
Materials Science ◽  
Causal Relationships ◽  
Structure And Properties ◽  
Compositional Field ◽  
Artificial Neural ◽  
Design Materials

Abstract The traditional aim of materials science is to establish the causal relationships between composition, processing, structure, and properties with the intention that, eventually, these relationships will make it possible to design materials to meet specifications. This paper explores another approach. If properties are related to structure at different scales, there may be relationships between properties that can be discerned and used to make predictions so that knowledge of some properties in a compositional field can be used to predict others. We use the physical properties of the elements as a dataset because it is expected to be both extensive and reliable and we explore this method by showing how it can be applied to predict the polarizability of the elements from other properties.

Artificial Neural Networks Used for Development Prediction of State-of-the-Art Surface Engineering Areas

2015 ◽  
Vol 220-221 ◽  
pp. 785-789
Author(s):  
Anna Danuta Dobrzańska-Danikiewicz ◽  
Jacek Trzaska ◽  
Agnieszka Sękala ◽  
Adam Jagiełło
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Computational Models ◽  
Surface Engineering ◽  
Materials Science ◽  
Added Value ◽  
Future Research ◽  
Processing Technologies ◽  
Future Events ◽  
Artificial Neural

The paper presents new, possible applications of artificial neural networks in the field of materials science and material engineering in relation to other artificial intelligence methods known and applied in this area. The most recent simulation experiments, the exemplary results of which are presented in this paper, point out that the scope of the existing applications of artificial neural networks can be extended to encompass new areas related to prediction of development of materials treatment and processing technologies. The goal of such research is to focus, intentionally, the areas of future research and investments on the most promising areas likely to yield the highest added value in the future together with mitigating a risk relating to such a process. The computational models created were used for creating multi-variant probabilistic scenarios of future events based on heuristic independent variables acquired in the process of multi-stage expert surveys. Dependencies were determined, in particular, between the probability of occurrence of alternative macro-scenarios of future events and the development of the relevant thematic areas of M1–M7 and P1–P7.

Sign in / Sign up

Export Citation Format

Share Document