Evolving flexible beta basis function neural tree using extended genetic programming & Hybrid Artificial Bee Colony

Abstract

tIn this paper, a new hybrid learning algorithm is introduced to evolve the flexible beta basis functionneural tree (FBBFNT). The structure is developed using the Extended Genetic Programming (EGP) and theBeta parameters and connected weights are optimized by the Hybrid Artificial Bee Colony algorithm. Thishybridization is essentially based on replacing the random Artificial Bee Colony (ABC) position with theguided Opposite-based Particle Swarm Optimization (OPSO) position. Such modification can minimizethe delay which might be lead by the random position, in reaching the global solution. The performanceof the proposed model is evaluated for benchmark problems drawn from time series prediction area andis compared with those of related methods.