Toward a Hybrid Approach of Primitive Cognitive Network Process and Particle Swarm Optimization Neural Network for Forecasting

Forecasting by artificial neural network is a popular approach in recent years. This paper proposes a new hybrid approach PCNP-PSONN which combines the Primitive Cognitive Network Process (PCNP) and Particle Swarm Optimization Neural Network (PSONN) for forecasting. PCNP is a rectified approach of the Analytic Hierarchy Process (AHP), to quantify the influence of factors, whilst Particle Swarm Optimization has been used for optimizing the neural network by improving the learning efficiency. The combination of PCNP and PSONN, PCNP-PSONN, can increase accuracy of network through selection of high influenced factors.