The advancement of data-driven approaches and methods has paved the way to complement and aid in overcoming the limitations of the traditional drug discovery processes. The utilization of machine learning–based approaches has impacted every step of drug discovery, including target identification, prioritization of inhibitors from large chemical space, and assessment of pharmacokinetic and pharmacodynamic properties. These methods not only facilitate the discovery of new chemical entities but can also be applied to identify new disease indications for existing drugs. This present chapter aims to provide insight into the application of machine learning–based and deep learning–based approaches developed for drug repurposing. The chapter discusses the application of AI methods toward the identification and prioritization of drug-like compounds through knowledge graph–based and deep learning–based ADMET methods. Moreover, the chapter discusses various resources, methods, and case studies for specific disease conditions, discussing a few success stories.