Abstract: This paper proposes a new landslide displacement prediction model. The model is based on deep learning which considers the hysteresis effect of landslide displacement. There are limitations in the application of existing landslide prediction models, due to the deformation and development characteristics of rainfall-induced landslide in southeastern China, in which there are plenty of mountains and hills. Therefore, it is of great theoretical and practical significance to study the landslide prediction model for disaster prevention and reduction. The model is based on intelligent algorithm, combining with the characteristics of landslide deformation. Research selected the landslide displacement monitoring data from September 2019 to June 2022, in Mount Yao, Anxi, Fujian Province. The model adopted gray relational analysis, set pair analysis and optimization of deep extreme learning machine based on sparrow search algorithm. Results show that the mean absolute error, the mean absolute percentage error and the root mean square error（RMSE） of the proposed SSA-DELM are all lower than existing BP, SVM model. Moreover, model integrates with influence factors of landslide and hysteresis effect of water level displacement.