Several machine learning methods can be used to predict ground conditions ahead of TBMs with high accuracy. Ensemble methods have better ground condition prediction accuracy than other machine learning models evaluated. The classification system used in characterizing the ground condition affects the performance of the machine models. The prediction performance of the machine models is different in soils and rocks of different lithologies. There have been significant advances in the use of both unsupervised and supervised machine learning (ML) methods to predict the ground condition or rock mass class ahead of tunnel boring machines (TBMs).