Machine Learning Driven New Material Discovery
New materials can bring tremendous technology and application progress. But now the commonly used trierand-error method can’t meet the need today. Now, a newly proposed idea using machine learning to explore new materials is becoming popular. In this paper, we review this research paradigm of applying machine learning in material discovery, including data preprocessing, feature engineering, machine learning algorithms and cross-validation procedures. Furthermore, we propose to assist traditional DFT calculation with this idea for material discovery. Many experiments and literatures have shown the great effect and prospect of this idea. It’s now showing its potential and advantages in property prediction, material discovery, inverse design, corrosion detection and many other aspects in life.