Abstract
Conference Title: 2014 14th International Conference on Hybrid Intelligent Systems (HIS) Conference Start Date: 2014, Dec. 14 Conference End Date: 2014, Dec. 16 Conference Location: Kuwait, Kuwait In this paper, we apply the Beta Basis Function Neural Network (BBFNN) trained with cuckoo search (CS) for time series predictions. The cuckoo search algorithm optimizes the network parameters. In order to evaluate the effectiveness of the proposed method, we have carried out some experiments on four data sets: Mackey Glass, Lorenz attractor, Henon map and Box-Jenkins. We give also simulation examples to compare the effectiveness of the model with the other known methods in the literature. The results show that the CS-BBFNN model produces a better generalization performance.