Accurate prediction of the land reclamation settlement is a long-standing problem. Numerous empirical, analytical, and numerical methods have been proposed to solve this problem, but they require large domain knowledge and experiments for calibrating parameters. This study aims to apply deep learning algorithm, long short-term memory (LSTM), to predict the long-term settlement of land reclamation settlement and apply the LSTM-based model to land reclamations of Kansai International Airport (KIA) and then Chek Lap Kok Airport (CLKA). The LSTM-based model is first trained based on the historical settlement data of KIA, and its optimum configurations are determined in a trial-and-error fashion. The well-developed model is subsequently applied to predict the settlement of an embankment in CLKA. The results indicate that the LSTM-based model can accurately capture the long-term settlement of KIA. The developed model exhibits excellent generalization ability and can be directly applied to the other project, i.e. CLKA with the predicted settlement in good agreement with measured data. The model is verified generic and can thus be applied to similar projects. A Graphical User Interface is finally developed to make the LSTM-based model available for engineering practice.