Intelligent forecasting system based on integration of electromagnetism-like mechanism and fuzzy neural network

作者：

Highlights：

• We integrate a new meta-heuristic algorithm (EM) into the FNN training process.

• The use of fuzzy ranking eliminated redundant weights and improved performance over other FNN models.

• The performance of EM-FNN is better than that of the FNN and GA-FNN models.

摘要

•We integrate a new meta-heuristic algorithm (EM) into the FNN training process.•The use of fuzzy ranking eliminated redundant weights and improved performance over other FNN models.•The performance of EM-FNN is better than that of the FNN and GA-FNN models.

论文关键词：Fuzzy neural networks,Electromagnetism-like mechanism,Back propagation algorithm,Weights elimination

论文评审过程：Available online 14 November 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.11.007