Open Access

REVIEW

Data-Driven Materials Science Using Machine Learning and Computational Modeling

Manjodh Kaur¹, Princy Randhawa^2,*, Jitendra Jaiswal², Deepak Dubal³, Ravindra N. Bulakhe^4,5, Deepanraj Balakrishnan⁶, Nithesh Naik^7,*

1 Department of Chemistry, School of Engineering, Dayananda Sagar University, Harohalli, Bengaluru, Karnataka, India
2 Computer Science and Engineering (AI&ML), School of Engineering, Dayananda Sagar University, Harohalli, Bengaluru, Karnataka, India
3 Faculty of Science, School of Chemistry & Physics, Queensland University of Technology, Brisbane, QLD, Australia
4 Center for 2D Quantum Heterostructures, Institute for Basic Science (IBS), Sungkyunkwan University (SKKU), Suwon, Republic of Korea
5 Symbiosis Centre for Nanoscience and Nanotechnology, Symbiosis International (Deemed University), Pune, India
6 Department of Mechanical Engineering, College of Engineering, Prince Mohammad Bin Fahd University, Al-Khobar, Saudi Arabia
7 Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India

* Corresponding Authors: Princy Randhawa. Email: ; Nithesh Naik. Email:

Computers, Materials & Continua 2026, 88(2), 4 https://doi.org/10.32604/cmc.2026.079503

Received 22 January 2026; Accepted 23 April 2026; Issue published 15 June 2026

Abstract

This review emphasizes the growing role of artificial intelligence (AI) in transforming the materials discovery process into a data-driven and autonomous approach. It systematically traces the evolution of scientific paradigms in materials science and examines how machine learning, generative models, and AI agents are revolutionizing the design, screening, and optimization of materials. A key contribution is a detailed, step-by-step machine learning framework that guides researchers through data collection, preprocessing, feature engineering, model development, and validation, utilizing publicly available materials databases and computational tools. Additionally, the review discusses the latest advances in generative AI and autonomous research systems, highlighting their potential to enable inverse design and closed-loop experiments. It includes a tutorial case study on sodium-ion battery materials to demonstrate practical application in formation energy prediction via machine learning, along with comparisons to high-throughput screening accuracy using density functional theory (DFT). The article also addresses current challenges such as data limitations, model interpretability, and physics-based approaches. Overall, this publication serves as both a conceptual and practical guide for integrating AI into materials research, aiming to accelerate the discovery process and improve efficiency.

---

Keywords

---