Tutorial
Can you rebuild this microstructure?
Generate colored patterns, guess the component count, then watch PCA reconstruct the image.
Latest articles and tutorials
All articles →
Start here
SVD, explained with a small picture
Build a picture one layer at a time. Start with everyday comparisons, then try the interactive example.
Tutorial
PCA with small pixel images
Follow the pixel values through centering, principal components and reconstruction. Run the notebook in small steps.
Data analysis
What does PCA actually show?
Explore directions of variation and learn why a mathematical pattern needs a physical explanation.
Machine learning
A good fit is not always a good prediction
Change the model and the noise to compare fitting the observations with predicting the underlying trend.
Materials design
Why even a small alloy design space becomes large
Count possible compositions and processing choices with an interactive design-space example.
A place to begin
Start with a picture.
You do not need to know matrix algebra to begin. See how numbers form a small image, then learn how SVD and PCA describe its patterns.
- Understand SVD with a small pictureAdd numerical layers and see what they change.
- Explore what PCA showsLook at directions of variation in a dataset.
- Work through the PCA pixel tutorialTry the controls, then run the notebook step by step.
Why materials data?
The significance of learning data analysis for materials
Generally, most researchers (experimentalists) conduct innumerable experiments to reach the final target. In this process, we generate tons of data that involves enormous consumption of energy, resources, and, most importantly, time. The whole activity has a direct impact on our climate, too. Therefore, designing and performing relevant experiments by analyzing pre-existing data to predict new or improved materials via materials informatics and analytics is the need of the hour.
What I share here
Learn, try, and explore.
- Image processing using Python
- A step-by-step guide to materials data analysis via machine learning and deep learning
- Review of recent developments in the field of materials informatics
I will share the codes developed by me and also provide tutorials on how to use publicly available open source codes related to multiscale materials design and informatics.
Click on articles and tutorials to begin your adventure with materials data.
About Materials Data Explorer
I am Joyita Bhattacharya, PhD, a materials scientist and educator with expertise in microscopy, materials characterization, materials informatics, and machine learning. Through this website, I share tutorials, practical examples, and research-inspired articles that help students, researchers, and engineers apply modern data science techniques to real materials science problems.
Whether you are taking your first steps in materials informatics or looking to deepen your expertise, I hope these resources help you analyze data more effectively, interpret results with greater confidence, and discover new insights from materials datasets. Explore the tutorials to build practical skills, or browse the blog for applications, case studies, and discussions at the intersection of materials science, microscopy, and data analytics.