An Algorithm Research for Prediction of Extreme Learning Machines Based on Rough Sets

Abstract

As is known, the extreme learning machine (ELM) algorithm is a new learning algorithm for the single hidden layer feedforward neural networks (SLFNs). This paper combines rough sets theory with extreme learning machine and proposes a new prediction algorithm of extreme learning machines based on rough sets theory. Firstly, using the rough sets theory to do attribute reduction, and then using ELM to train and predict the new datasets. Verified by the final experimental results and data analysis we know that compared with the traditional ELM the proposed algorithm has higher prediction accuracy and better efficiency.