Application of Multivariate Data Analysis Techniques in Modeling Structure–Property Relationships of Some Superconductive Cuprates

Lehmus, K.; Karppinen, Maarit

doi:10.1006/jssc.2001.9283

Cited by 7 publications

(7 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most recent studies suggest that the chemical information could be successfully integrated with techniques of machine learning [8,[10][11][12][17][18][19]. A series of predictive models that explore quantitative relationships between critical temperature and physicochemical properties of materials have been reported in the literature [1,6,20,21]. One of the pioneering works directly attributes critical temperatures of 60 high-temperature superconductors to valence-electron numbers, orbital radii, and electronegativity [21].…”

Section: Introductionmentioning

confidence: 99%

“…One of the pioneering works directly attributes critical temperatures of 60 high-temperature superconductors to valence-electron numbers, orbital radii, and electronegativity [21]. Later, PCA and PLA were applied to predict TC for 1212 superconductive copper oxides [20]. Most recently, predictive and classification models were generated for more than 10,000 known superconductors using the RF, MLR, and gradient boosting techniques [1,6].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predictive Modeling of Critical Temperatures in Superconducting Materials

Sizochenko

Hofmann

2020

Molecules

View full text Add to dashboard Cite

In this study, we have investigated quantitative relationships between critical temperatures of superconductive inorganic materials and the basic physicochemical attributes of these materials (also called quantitative structure-property relationships). We demonstrated that one of the most recent studies (titled "A data-driven statistical model for predicting the critical temperature of a superconductor” and published in Computational Materials Science by K. Hamidieh in 2018) reports on models that were based on the dataset that contains 27% of duplicate entries. We aimed to deliver stable models for a properly cleaned dataset using the same modeling techniques (multiple linear regression, MLR, and gradient boosting decision trees, XGBoost). The predictive ability of our best XGBoost model (R2 = 0.924, RMSE = 9.336 using 10-fold cross-validation) is comparable to the XGBoost model by the author of the initial dataset (R2 = 0.920 and RMSE = 9.5 K in ten-fold cross-validation). At the same time, our best model is based on less sophisticated parameters, which allows one to make more accurate interpretations while maintaining a generalizable model. In particular, we found that the highest relative influence is attributed to variables that represent the thermal conductivity of materials. In addition to MLR and XGBoost, we explored the potential of other machine learning techniques (NN, neural networks and RF, random forests).

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Predictive Modeling of Critical Temperatures in Superconducting Materials

Sizochenko

Hofmann

2020

Molecules

View full text Add to dashboard Cite

show abstract

“…Numerous examples of the application of data mining for materials problems exists in the literature. [15][16][17][18][19][20][21][22] However, the use of a multivariate analysis with polymer systems has been more limited. [23][24][25][26][27] With PCA, the most important features of the FTIR spectra can be identified, and the peak shifts and non-symmetries in the peaks between the samples can be quickly determined.…”

Section: Introductionmentioning

confidence: 99%

Tracking Chemical Processing Pathways in Combinatorial Polymer Libraries via Data Mining

et al. 2009

View full text Add to dashboard Cite

Changes in the molecular structure and composition of interpenetrating polymer networks (IPNs) can be used to tailor their properties. While the properties of IPNs are typically different than polymer blends, a clear understanding of the impact of changing polymerization sequence on the physical properties and the corresponding molecular bonding is needed. To address this issue, a data mining approach is used to identify the change with polymerization sequence of tensile and rheological properties of acrylate-epoxy IPNs. The experimental approach used to study the molecular structure is high throughput Fourier transform infrared (FTIR) spectroscopy. Analysis of the FTIR spectra of IPNs synthesized with different polymerization sequences leads to an understanding of the molecular bonding responsible for the tensile and rheological properties. From the interpretation of the wavenumber bands and associated molecular bonds, we find that the polymerization sequence most affects hydrogen bonding and aromatic ring bond energies. This work defines the relationships between chemistry, structure, processing, and properties of the IPN samples. Changes in the molecular structure and composition of interpenetrating polymer networks (IPNs) can be used to tailor their properties. While the properties of IPNs are typically different than polymer blends, a clear understanding of the impact of changing polymerization sequence on the physical properties and the corresponding molecular bonding is needed. To address this issue, a data mining approach is used to identify the change with polymerization sequence of tensile and rheological properties of acrylate-epoxy IPNs. The experimental approach used to study the molecular structure is high throughput Fourier transform infrared (FTIR) spectroscopy. Analysis of the FTIR spectra of IPNs synthesized with different polymerization sequences leads to an understanding of the molecular bonding responsible for the tensile and rheological properties. From the interpretation of the wavenumber bands and associated molecular bonds, we find that the polymerization sequence most affects hydrogen bonding and aromatic ring bond energies. This work defines the relationships between chemistry, structure, processing, and properties of the IPN samples.

show abstract

“…PCA has been applied to various materials to study property‐structure relationships 20–25. This paper will build on the previous work by combining the elements of experimentation on IPN structure‐property‐processing relationships, combinatorial methods, and informatics to develop additional insights into the development and characterization of these complex materials.…”

Section: Introductionmentioning

confidence: 99%

Combinatorial Methods and Informatics Provide Insight into Physical Properties and Structure Relationships during IPN Formation

Nowers

Broderick

Rajan

et al. 2007

Macromol. Rapid Commun.

View full text Add to dashboard Cite

Combinatorial methods and informatics are applied to the study of complex property‐structure‐processing relationships during interpenetrating polymer network formation in an epoxy/acrylate system. Principal component analysis of a dataset that covers different compositions and process sequences successfully identifies the most unique samples as well as relationships between material properties. The relationships between material properties can be exploited in future investigations by allowing high throughput screening and as a guide for engineering materials. The use of combinatorial methods, high throughput screening, and informatics will lead to accelerated material design.magnified image

show abstract

Application of Multivariate Data Analysis Techniques in Modeling Structure–Property Relationships of Some Superconductive Cuprates

Cited by 7 publications

References 37 publications

Predictive Modeling of Critical Temperatures in Superconducting Materials

Predictive Modeling of Critical Temperatures in Superconducting Materials

Tracking Chemical Processing Pathways in Combinatorial Polymer Libraries via Data Mining

Combinatorial Methods and Informatics Provide Insight into Physical Properties and Structure Relationships during IPN Formation

Contact Info

Product

Resources

About