What is it about?
The project explores how combining computing and materials science brings new insights. Using data from 2,574 different materials, the focus was on predicting how the combination of elements in a material affects its ability to handle stress (known as shear modulus). A method was used to convert the material data into a form that computers can analyze, and different machine learning techniques were applied to make predictions. A deep learning model was also tested to see if it could improve the accuracy of the results.
Featured Image
Photo by Hunter Harritt on Unsplash
Why is it important?
Incorporating physics into machine learning aims to develop smarter algorithms that can maintain key physical properties and help align simulations with real-world experiments. This approach speeds up the discovery and creation of new materials by predicting their performance, leading to lower costs in product development. Machine learning also makes better use of existing networks and systems, fostering intelligent collaboration and new methods for teams to work together and create improved products.
Perspectives
I hope this project makes what might seem like a technical and abstract field—combining machine learning with materials science—feel more accessible and intriguing. The way we use advanced algorithms to predict material properties and improve material development isn't just a concern for scientists and engineers; it impacts many areas of technology and innovation that affect our daily lives. More than anything, I hope you find this exploration thought-provoking and see the exciting potential in the intersection of these fields.
Malavika G Prasad
Amrita Vishwa Vidyapeetham
Read the Original
This page is a summary of: Predicting shear modulus property using materials informatics, January 2024, American Institute of Physics,
DOI: 10.1063/5.0228632.
You can read the full text:
Contributors
The following have contributed to this page
