Title: A multi-factor prediction algorithm in big data computing environments

Authors: Hao Tang; Dawei Sun

Addresses: School of Information Engineering, China University of Geosciences, Beijing 100083, China ' School of Information Engineering, China University of Geosciences, Beijing 100083, China

Abstract: In big data environments, many problems become more huge and complex. This makes it harder for us to deal with the data and forecast the result. Traditional prediction algorithm is more suitable for less factor problem, and how to solve multi-factor problem is one of the major challenges. In this paper, a more accurate and effective algorithm was proposed based on the good performance of weight optimisation of PSO and generalisation ability of BPNN. Further, the convergence and feasibility of the combinational algorithm were analysed with actual data. The simulation revealed that with the increase of influencing factors, the BP neutral network optimised by PSO possessed a more rapid convergence rate of MSE and higher accuracy of network output values.

Keywords: data prediction; big data computing; multi-factor prediction; weight optimisation; generalisation; back-propagation ANNs; artificial neural networks; particle swarm optimisation; PSO; simulation.

DOI: 10.1504/IJCSM.2016.078735

International Journal of Computing Science and Mathematics, 2016 Vol.7 No.4, pp.312 - 322

Received: 07 Apr 2016
Accepted: 26 Apr 2016
Published online: 01 Sep 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article