Aiming at the characteristics of small sample data and high feature dimension in risk prediction of high-rise building project duration,a risk prediction model of high-rise building project duration was proposed based on least squares support vector machine(LSSVM) optimized by random forest(RF)algorithm and particle swarm optimization(PSO)algorithm.RF algorithm with obvious advantages in feature selection was used to select the best feature subset.PSO algorithm was used to optimize the regularization parameters and kernel function parameters of LSSVM.The precision rate,recall rate and F1m value were used to verify and evaluate the predictive performance of the proposed model.The research results show that the average precision rate reaches 93.71%,and the averagere call rate reaches 94.04% to predict the risk of high-rise building project duration using the proposed model. The proposed model can accurately predict the risk level of high-rise building project duration,improves the risk prediction method of high-rise building project duration,and the predicted results would provide a reference for risk control of high-rise building project duration.