This research aims at making use of advanced Machine Learning Algorithms (MLAs) with a view to developning a precise geoid model for Egypt. Being an equipotential surface of the earth's gravity field, geoid plays a crucial role in various geodetic applications. Throughout this study, state-of-the-art Machine Learning Algorithms are employed to address the limitations of conventional geoid modeling approaches. The research methodology involves evaluating the performance of eight Global Geopotential Models(GGMs), namely EGM2008, EIGEN-6C, EIGEN-6C2, EIGEN-6C4, EIGEN-6C3stat, SGG-UGM-1, XGM2019e_2159 and SGG-UGM-2 to choose the suitable GGM that for the study area, i.e. Egypt. MLAs, such as Linear Regression, Support Vector Machine, Random Forest, and Extra Trees, are then applied to train a model capable of determinig the intricate relationships between the input features and the geoid undulations. The study findings conclude that XGM2019e_2159 emerges as the optimal GGM for Egyptian territories, since it has yielded a standard deviation of 0.36 m. Notable enhancements in the local geoid model are observed with the application of the Extra Trees algorithm, which has yielded a standard deviation of 0.11 m.
 
 
Special Issue of AEIC 2024 (Civil Engineering  Session)