In materials informatics, combining cheminformatics with machine learning is a powerful way to accelerate novel materials design. This chapter provides an easy-to-understand introduction to machine learning techniques designed for predicting outcomes in materials informatics. Beginning with foundational principles such as supervised and unsupervised learning, the chapter explains important algorithms for predictive tasks, such as regression, classification, and clustering. Focusing on real-world use, the chapter covers how to evaluate models, choose the best features, and handle unique data challenges in materials datasets. By explaining machine learning clearly in materials informatics, this chapter helps readers use computational tools effectively to create innovative materials.