Machine learning, either classical or quantum, has been employed to solve extensively diverse problems in materials science. They capture the complicated component-structure-property relationship and accelerate the traditional simulation methods. We will introduce the frequently used machine learning algorithms and summarize their important applications, providing representative examples.