基于智能算法的价格预测模型探究

摘要

文中主要是建立基于智能算法的价格预测模型，解决商品价格预测的难题。首先通过德尔菲技术进行专家意见征询，确定影响商品价格的因子；然后数据化处理影响因子后，利用拉依达准则剔除异常数据；再以影响因子为输入量，采用BP神经网络算法建立商品预测模型；采用聚类分析算法建立预测模型与之对比。以玫瑰鲜切花为例建立价格预测模型，实验结果表明：该商品价格预测模型规避了单纯BP神经网络算法的缺陷，具有预测商品价格的普遍性、实用性。%In order to estimate commodity prices,the purpose in this paper is to establish the forecasting price model. Firstly,through ex-pert consulting by the Delphi technique,the factors influencing commodity prices are determined. Secondly,after processing the factors, data are got,and abnormal data of them are eliminated according to Pauta criterion. Thirdly,with the influencing factors as input data,es-tablish the forecasting commodity prices model by means of the BP neural network algorithm. Finally,this algorithm is compared with the clustering analysis algorithm applied in the forecasting model. This forecasting model has been applied to predict the rose cut flowers. The experimental results show that this forecasting model can avoid defects of traditional BP algorithm,and has certain generality and practi-cality.