Development of forecasting model of fashion sales based on Back propagation neural network and correlation analysis

Abstract

The result of operational decisions of fashion should be evaluated accuracy. In order to forecast accuracy, an improved forecasting algorithm, which combines correlation analysis, artificial neural network and the other computation skills, is proposed in this paper. The analysis of practical situations indicates that this improved algorithm can generate reasonably good forecasting quickly under the condition of limited data. We have compared our method with back propagation neural network combined with genetic algorithm. The result shown that our developed method can forecast the fashion sales with better results.