Under the "dual carbon" background, the development of high-performance energy and environmental catalysis materials is of great significance for promoting energy clean transformation and environmental pollution control. The traditional research and development mode of catalysts mainly relies on experimental and trial-and-error methods, which to a large extent cannot meet the research and development needs of efficient catalysts in emerging energy and environmental fields. The rapid development of data science technologies such as machine learning is expected to bring about a paradigm shift in catalyst research and development. By using machine learning methods to quickly screen high-performance energy and environmental catalysis materials using experimental or computational data, the limitations of traditional trial-and-error methods could be overcome, and the problem of low efficiency and high cost in catalyst research and development could be solved. This article reviewed the main processes and research progress of machine learning methods in the development of energy and environmental catalysis materials from the perspective of active-sites prediction,catalysts screening, morphology design and reaction mechanism revelation, and the ML methods construction corresponding to various training data acquisition and their applications in the catalytic research. We also discussed the future direction of this method in the catalysis field, in order to provide perspective and promote its application in the energy and environmental fields.