Improved Extreme Learning Machine Model for Grop Yield Prediction
WU Yaodong
Abstract：To improve the accuracy and timeliness of grop yield prediction,a forecasting algorithm of grain yield based on growing extreme learning machine(GELM)is proposed,whose theoretical framework is extreme learning machine(ELM). In order to solve the problem of hidden nodes number L optimization for ELM,under the situation of L increasing,the recursive formula of output weight generalized inverse matrix is deduced. The deduced equation avoids the problem that output weight generalized in?verse matrix is calculated repeatedly under different L,which reduces ELM computation load. Then the GELM is presented. Finally the grain yield prediction flow is given. Taking 1960-2015 year grain production of China as experiment dataset,testing results show that,comparing with ELM and support vector machine(SVM),GELM prediction accuracy is higher than SVM,and is close to ELM. However,GELM time consuming is much lower than ELM and SVM.
Keywords：extreme learning machinetime series predictiongeneralized inverse matrixgrain productionneural network
Publication Date：2019-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 1283-1286,1302 )