propnet: A Knowledge Graph for Materials Science

Moreau, David; Horton, Matthew K.; Montoya, Joseph H.; Legaspi, Christian M.; Dwaraknath, Shyam; Tshitoyan, Vahe; Jain, Anubhav; Persson, Kristin A.

doi:10.1016/j.matt.2019.11.013

Cited by 47 publications

(23 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For integration with experiments, one could imagine a 'meta-decision-tree' or a Bayesian algorithm in conjunction with an automated characterization probe in real time, scanning energy ranges with high expected information value until sufficient signal is achieved to determine structural properties of interest with reasonably high accuracy, analogously to an active learning loop as used in other contexts for ML and materials science [1,12,71,72].…”

Section: Discussionmentioning

confidence: 99%

Random Forest Machine Learning Models for Interpretable X-Ray Absorption Near-Edge Structure Spectrum-Property Relationships

Torrisi

Carbone²,

Rohr³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

X-ray absorption spectroscopy (XAS) produces a wealth of information about the local structure of materials, but interpretation of spectra often relies on easily accessible trends and prior assumptions about the structure. Recently, researchers have demonstrated that machine learning models can automate this process to predict the coordinating environments of absorbing atoms from their XAS spectra. However, machine learning models are often difficult to interpret, making it challenging to determine when they are valid and whether they are consistent with physical theories. In this work, we present three main advances to the data-driven analysis of XAS spectra: we demonstrate the efficacy of random forests in solving two new property determination tasks (predicting Bader charge and mean nearest neighbor distance), we show that multiscale featurization can elucidate the regions and trends in spectra that encode various local properties, and we address the effect of normalization on model interpretability. The multiscale featurization transforms the spectrum into a vector of polynomial-fit features, and is contrasted with the commonly-used "pointwise" featurization that directly uses the entire spectrum as input. We find that across thousands of transition metal oxide spectra, the relative importance of features describing the curvature of the spectrum can be localized to individual energy ranges, and we can separate the importance of constant, linear, quadratic, and cubic trends, as well as the white line energy. This work has the potential to assist rigorous theoretical interpretations, expedite experimental data collection, and automate analysis of XAS spectra, thus accelerating discovery of new functional materials.

show abstract

Section: Discussionmentioning

confidence: 99%

Random Forest Machine Learning Models for Interpretable X-Ray Absorption Near-Edge Structure Spectrum-Property Relationships

Torrisi

Carbone²,

Rohr³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several studies about the relationship between the energy gap and the refractive index have been proposed for semiconductors and examined in the past, yielding various theories in this field [27][28][29]. There has been renewed interest in these studies in recent years [30][31][32][33][34][35][36][37]. While several manuscripts have reported on the studies of the energy gap and refractive index of perovskites [38][39][40][41], Blessing N. Ezealigo et.al.…”

Section: Introductionmentioning

confidence: 99%

“…A comprehensive study of the fundamental properties such as the energy gap and refractive index is of paramount importance for the study of materials, in particular perovskites, since they are the basis for determining their applications in electronics and photonics. Furthermore, computational frameworks in materials science such as "propnet" [34] require pre-knowledge of the database of these material properties. As materials informatics begins to grow, investigations such as these that relate two fundamental macroscopic properties will pave the way for new applications of perovskites.…”

Section: Introductionmentioning

confidence: 99%

Energy Gap-Refractive Index Relations in Perovskites

Lamichhane

Ravindra

2020

Materials

View full text Add to dashboard Cite

In this study, the energy gap-refractive index relations of perovskites are examined in detail. In general, the properties of perovskites are dependent on the structural reorganization and covalent nature of their octahedral cages. Based on this notion, a simple relation governing the energy gap and the refractive index is proposed for perovskites. The results obtained with this relation are in good accord with the literature values and are consistent with some well-established relations.

show abstract

“…Despite the high expectations around materials informatics, even cutting-edge prediction models are not yet able to integrate big data from materials science, due to the lack of general knowledge in this field [6][7][8][9][10][11][12][13] . A number of models predict a variety of material parameters, including physical and chemical properties, structures, and spectroscopic responses [6][7][8][9][10][11][12][13][14][15][16][17][18] . The recent development of data mining techniques from scientific literature is also helpful to increase the number of databases and to enhance prediction accuracy 14,15,18 .…”

mentioning

confidence: 99%

“…1). Graph approaches have been employed to analyze the relationships of atom-connections, chemical features, and reactions 6,16,17 . In this study, we extended the approach to train a neural network with the general phenomena of science, which are expressed by graphs.…”

mentioning

confidence: 99%

Integrating multiple materials science projects in a single neural network

Hatakeyama‐Sato

Oyaizu

2020

Commun Mater

View full text Add to dashboard Cite

In data-intensive science, machine learning plays a critical role in processing big data. However, the potential of machine learning has been limited in the field of materials science because of the difficulty in treating complex real-world information as a digital language. Here, we propose to use graph-shaped databases with a common format to describe almost any materials science experimental data digitally, including chemical structures, processes, properties, and natural languages. The graphs can express real world's data with little information loss. In our approach, a single neural network treats the versatile materials science data collected from over ten projects, whereas traditional approaches require individual models to be prepared to process each individual database and property. The multitask learning of miscellaneous factors increases the prediction accuracy of parameters synergistically by acquiring broad knowledge in the field. The integration is beneficial for developing general prediction models and for solving inverse problems in materials science.

show abstract

propnet: A Knowledge Graph for Materials Science

Cited by 47 publications

References 42 publications

Random Forest Machine Learning Models for Interpretable X-Ray Absorption Near-Edge Structure Spectrum-Property Relationships

Random Forest Machine Learning Models for Interpretable X-Ray Absorption Near-Edge Structure Spectrum-Property Relationships

Energy Gap-Refractive Index Relations in Perovskites

Integrating multiple materials science projects in a single neural network

Contact Info

Product

Resources

About