Rental Volume Prediction of Shared Bicycles Based on ISSA-Stacking Ensemble Learning
To solve the problem of imbalance between supply and demand of shared bicycles,a prediction model for shared bicycle rental volume based on ISSA-Stacking algorithm is proposed by combining Stacking algorithm and improved sparrow search algorithm(ISSA).Firstly,the important features are selected by using correlation analysis method and light gradient boosting machine.Then,multiple heterogeneous regression prediction models are established and the key hyperparameters of each model are optimized by ISSA algorithm,and the global search ability and convergence speed of ISSA are improved by introducing elite reverse learning strategy and adaptive population scale factor.Finally,the models are integrated by using the ensemble learning idea of Stacking algorithm.In the experiment,the shared bicycle travel data in Washington area of the United States are used to predict the rental volume.Through comparative analysis,it is verified that the proposed integrated model has better prediction accuracy than the single model in the prediction of shared bicycle rental volume.
万方数据
维普
