A new study introduced a machine learning-powered clustering model that incorporates both basic features and target properties, successfully grouping over 1,000 inorganic materials.