Using a mixture of experts’ approach to solve the forecasting task

Abstract

The forecasting problem appears frequently in the aviation industry (demand forecasting, air transport movement forecasting, etc.). In this article, a new approach based on multiple neural networks of different topologies is introduced. An algorithm was tested on real data and showed better results compared to several other methods. This shows its suitability for further usage in aviation forecasting tasks.

First published online: 02 Oct 2014

Keyword : artificial neural network, forecasting method, mixture of experts, mean square prediction error, group method of data handling, training set