Abstract: With the advantages of high forming efficiency and low cost, directed energy deposition-arc technology has become an indispensable part of metal additive manufacturing technology. However, directed energy deposition-arc has poor stability, and requires a lot of manual intervention in the manufacturing process. There is low degree of intelligence and other challenges in directed energy deposition-arc. The analysis method of directed energy deposition-arc process based on the machine learning is of great significance to promote the intelligence of directed energy deposition-arc technology. Mainly from three aspects of directed energy deposition-arc design, process monitoring, and the relationship between process parameters and the performance of formed parts, the progress on the research and application of machine learning methods in the above three aspects was summarized. On this basis, the challenges and development prospects of the field were emphasized.