New algorithm for predicting ensemble aggregation fusion based on fuzzy error model and local neighbor revision
Ats Lab · Yiqi yibiao xuebao · 2011
Prediction models have randomness and sample data have uncertainty and deformity.Ensemble method can reduce the prediction error caused by randomness and uncertainty.This article proposes a method,which uses a revised fuzzy RBF neuron network(FRNN) algorithm to produce a neighborhood revision fusion of ensemble;an FRNN error-model is pre-processed to represent the runtime performance for each prediction model.In prediction,with the output of error model the ensemble aggregation method is used to guide a weight-bias fusion,and multi-prediction-models are aggregated in run time and the overall prediction output is given.The proposed method chooses a larger weight for a better prediction model,eliminates the local prediction error bias,and pre-processes all necessary error models before on-line prediction.Results show that the method has better performance than ordinary models and has good runtime efficiency,and good popularization and application value.