The geomagnetic data plays a important role in understanding the evolutionary process of Earth’s magnetic field, as it provides necessary information for near-surface exploration, unexploded explosive ordnance detection, and so on. To reconstruct the geomagnetic data, this project presents a geomagnetic data reconstruction method based on machine learning techniques. The traditional linear approaches are prone to time inefficiency and involves high labor cost, while the proposed approach has a significant improvement. In this project, three classic machine learning models, support vector machine, random forests, and gradient boosting were built. And, a deep learning algorithm, recurrent neural network, was explored to further improve the performance. The proposed learning methods were used to specify a continuous regression hyperplane from a training data. The specified regression hyperplane is a mapping of the relation between the missing data and the surrounding intact data. Then, the trained method, were used to build the missing geomagnetic data for validation, and they can be used for reconstructing further collected new field data. Finally, numerical experiments were derived. The results shows that the performance of our proposed methods was more accurate in comparison with the traditional linear learning method, as the reconstruction accuracy was increased by approximately 10%∼20%.