Distilling a Materials Synthesis Ontology

Kim, Edward; Huang, Kevin; Kononova, Olga; Ceder, Gerbrand; Olivetti, Elsa

doi:10.1016/j.matt.2019.05.011

Cited by 36 publications

(35 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth noting that the correlation is stronger when the NLP model is applied to extract synthesis steps rather than materials entities. This can be explained with the fact that the context of a sentence defining a synthesis action is more ambiguous than that for materials terms ( Kim et al., 2019 ). This complexity stresses the need to improve the general NLP tools to deal with scientific text.…”

Section: Challenges and Caveats Of The Text-mining-driven Researchmentioning

confidence: 99%

Opportunities and challenges of text mining in materials research

Kononova

Huo

et al. 2021

iScience

Self Cite

View full text Add to dashboard Cite

Research publications are the major repository of scientific knowledge. However, their unstructured and highly heterogenous format creates a significant obstacle to large-scale analysis of the information contained within. Recent progress in natural language processing (NLP) has provided a variety of tools for high-quality information extraction from unstructured text. These tools are primarily trained on non-technical text and struggle to produce accurate results when applied to scientific text, involving specific technical terminology. During the last years, significant efforts in information retrieval have been made for biomedical and biochemical publications. For materials science, text mining (TM) methodology is still at the dawn of its development. In this review, we survey the recent progress in creating and applying TM and NLP approaches to materials science field. This review is directed at the broad class of researchers aiming to learn the fundamentals of TM as applied to the materials science publications.

show abstract

Section: Challenges and Caveats Of The Text-mining-driven Researchmentioning

confidence: 99%

Opportunities and challenges of text mining in materials research

Kononova

Huo

et al. 2021

iScience

Self Cite

View full text Add to dashboard Cite

show abstract

“…BIG-MAP is developing an battery ontology, BattINFO, [30] which is based upon the European Materials Modelling Ontology community standard [31] and initiatives from other parts of the discovery cycle, for example, materials syntheses. [32] This will help ensure interoperability and be facilitated by open data archives and platforms such as the Materials Cloud [33] that allow users to store, explore, and share data in curated and well-defined formats that can be accessed from other platforms and for multiple purposes. By combining this with standardized and unified experimental protocols for battery testing and characterization that are currently under development, for example, in BATTERY 2030+, Batteries Europe and leading scientific journals in the field like Journal of Power Sources and Joule, it will be possible to develop autonomous workflows that can acquisition and utilize directly comparable and interoperable experimental and computational data.…”

Section: A Holistic Infrastructure For Autonomous Battery Discoverymentioning

confidence: 99%

Toward Better and Smarter Batteries by Combining AI with Multisensory and Self‐Healing Approaches

Vegge¹,

Tarascon

Edström

2021

Advanced Energy Materials

View full text Add to dashboard Cite

With an exponentially growing demand for rechargeable batteries, the development of new ultra‐performant, fully scalable, and sustainable battery technologies and materials must be accelerated. Creating a holistic, closed‐loop infrastructure for materials discovery, manufacturing, and battery testing that utilizes a common data infrastructure and autonomous workflows to bridge big data from all domains of the battery value chain, can pave the way for a transformative reduction in the required time to discovery. By embedding multisensory and self‐healing capabilities in future battery technologies and integrating these with AI and physics‐aware machine learning models capable of predicting the spatio‐temporal evolution of battery materials and interfaces, it will, in time, be possible to identify, predict and prevent potential degradation and failure modes. This will facilitate enhanced battery quality, reliability, and life, for example, by preemptively changing the battery charging conditions or releasing self‐healing additives from the separator membrane, akin to preemptive medicine, and form the basis for inverse design of new battery materials, interfaces, and additives. The large‐scale and long‐term European research initiative BATTERY 2030+ seeks to make this longer‐than ten‐year vision a reality through the development of a versatile and chemistry neutral “Battery Interface Genome—Materials Acceleration Platform” infrastructure (BIG‐MAP).

show abstract

“…Therefore, any predictive ML/AI model-as outlined in Challenges 5 and 6, for example-would benefit enormously from the inclusion of synthesis parameters as a feature. In addition, the extracting of synthesis protocols 47,83,84 can provide deeper insights into the literature by answering some questions such as: (i) what are the methods commonly used to prepare a glass?, (ii) what method should be used for a specific glass, (iii) what is the effect of a method on a property, (iv) what method should be adopted to obtain a target property, to name a few. One way to address this issue is to develop a consistent dataset with the same protocol followed throughout, as mentioned in Challenge 1.…”

Section: Challenge 16: Extracting the Synthesis Protocols From The Literaturementioning

confidence: 99%

Artificial intelligence and machine learning in glass science and technology: 21 challenges for the 21^st century

Ravinder

Venugopal

Bishnoi

et al. 2021

Int J of Appl Glass Sci

View full text Add to dashboard Cite

The progress of human civilization has always been closely associated with the discovery of new materials. This is probably why the tripartite classification of historical periods is also based on materials-stone, bronze, and iron age.Beyond these materials, there are several others which have significantly improved the quality of human life, namely, steel, aluminum, glass, plastics, the latest in the list being nanomaterials. Among these materials, glasses hold a unique place in human lives, considering their applications ranging from everyday glass utensils and kitchen-wares to

show abstract

Distilling a Materials Synthesis Ontology

Abstract: Methods sections adopt a common practice of past-tense narrative using passive voice. Here, we discuss issues with current and historical writing conventions in materials science literature and propose a structured way to facilitate reproducibility, clarity, and machine readability.

Cited by 36 publications

References 11 publications

Opportunities and challenges of text mining in materials research

Opportunities and challenges of text mining in materials research

Toward Better and Smarter Batteries by Combining AI with Multisensory and Self‐Healing Approaches

Artificial intelligence and machine learning in glass science and technology: 21 challenges for the 21^st century

Contact Info

Product

Resources

About

Distilling a Materials Synthesis Ontology

Abstract: Methods sections adopt a common practice of past-tense narrative using passive voice. Here, we discuss issues with current and historical writing conventions in materials science literature and propose a structured way to facilitate reproducibility, clarity, and machine readability.

Cited by 36 publications

References 11 publications

Opportunities and challenges of text mining in materials research

Opportunities and challenges of text mining in materials research

Toward Better and Smarter Batteries by Combining AI with Multisensory and Self‐Healing Approaches

Artificial intelligence and machine learning in glass science and technology: 21 challenges for the 21st century

Contact Info

Product

Resources

About

Artificial intelligence and machine learning in glass science and technology: 21 challenges for the 21^st century