next up previous
Next: Introduction

Parallel Back-Propagation for Sales Prediction on Transputer Systems

Frank M. Thiesing, Ulrich Middelberg, Oliver Vornberger
University of Osnabrück, Dept. of Math./Comp. Science
D-49069 Osnabrück, Germany


In this paper artificial neural networks are adapted to a short term forecast for the sale of articles in supermarkets. The data is modelled to fit into feedforward multilayer perceptron networks that are trained by the back-propagation algorithm. For enhancement this has been parallelized in different manners. One batch and two on-line training variants are implemented on parallel Transputer-based PARSYTEC systems: a GCel with T805 and a GC/PP with PowerPC processors and Transputer communication links. The parallelizations run with both the runtime environments PARIX and PVM.

Fri Jun 30 13:29:58 MET DST 1995