Materials science is an interdisciplinary field that aims to understand the physical basis of materials behaviour, optimize materials currently in use, and design new materials. The field of materials science has seen major revolution in terms of big data explosion and availability of large repositories, leading to the onset of new domain called as material informaticsMaterial informatics. Materials informatics incorporates machine learningMachine Learning (ML) based algorithms to screening materials quickly, generating tailored materials, identifying complex correlations to optimize the materials to name a few. Machine learning is a revolutionary technique that deftly extracts hidden patterns from data into insightful knowledge. In this chapter, we provide an overview of different machine learning approaches and applications in materials science, with emphasis on its ability to spur materials innovation, provide fresh perspectives, and solve both research and industry related problems. Different machine learningMachine Learning (ML) algorithms like artificial neural network, support vector machineSupport Vector Machine (SVM), decision treesDecision Trees (DT), deep learning modelsDeep learning models, along with emerging technologies like the explainable artificial intelligence (AI) methods have been discussed. Applications of machine learning based models offer promising direction for future research and development in materials science, which might have a significant influence on how materials innovation and research are conducted in the future.