A Broad Learning Ensemble System Using Bagging for Typhoon Trajectory Forecasting
Fengyun Hao, Jian Jin · 2022 2nd International Conference on Consumer Electronics and Computer Engineering (ICCECE) · 2022
The prediction of typhoon trajectory is a very important work, as typhoon is one of nature disasters. This paper builds a broad learning ensemble system by bagging to predict the typhoon trajectory in the South China Sea. Firstly, the prediction ability of single Broad Learning System is proved by our experiments. But considering the defect of instability, 10 Broad Learning System models are aggregated with bagging method in the framework of ensemble learning, to predict the typhoon trajectory. The experiments show that the mean distance error of Broad Learning Ensemble System is reduced by 17.86% compared with the stepwise regression model, while the stepwise regression model as classic model, is widely used in typhoon prediction practice.