Materials Knowledge Systems in Python - A Data Science Framework for Accelerated Development of Hierarchical Materials

Published

March 15, 2017

Author(s)

David Brough, Daniel Wheeler, Surya R. Kalidindi

Abstract

There is a critical need for customized analytics that take into account the stochastic nature of materials internal structure at multiple length scales in order to extract relevant and transferable knowledge. Data driven Process-Structure-Property (PSP) linkages provide systemic, modular and hierarchical framework for community driven curation of materials knowledge, and its transference to design and manufacturing experts. The Materials Knowledge Systems in Python project (PyMKS) is the first open source materials data science framework that can be used to create high value PSP linkages for hierarchical materials that can be leveraged by experts in materials science and engineering, manufacturing, machine learning and data science communities. This paper describes the main functions available from this repository, along with illustrations of how these can be accessed, utilized, and potentially further refined by the broader community of researchers.

Citation

Jom-Journal of the Minerals Metals & Materials Society

Pub Type

Journals

Keywords

Materials Knowledge Systems, Hierarchical Materials, Multiscale Materials, Python, Scikit-learn, NumPy, SciPy, Machine Learning

Citation

Brough, D. , Wheeler, D. and Kalidindi, S. (2017), Materials Knowledge Systems in Python - A Data Science Framework for Accelerated Development of Hierarchical Materials, Jom-Journal of the Minerals Metals & Materials Society (Accessed May 3, 2026)

Additional citation formats