基于数据挖掘的商品销售预测分析

Commodity Sales Forecasting Based on Data Mining

为了优化企业商品销售决策方案,提高商品销售预测的准确率,采用数据挖掘的方法对商品销售预测分析。首先对商品销售预测基本流程和现有预测算法进行归纳总结,接着提出了基于数据挖掘的商品销售预测框架,并对决策树及贝叶斯网络算法数学模型、关联规则进行阐述,最后运用实例仿真,对商品的一季度的销售额与利润额进行预测分析,实验表明,预测数据与实际数据基本一致,预测效果良好,具有一定的研究价值。

In order to optimize the decision scheme of commodity sales and improve the accuracy of sales forecasting, this paper adopts the way of data mining. It firstly summarizes the basic procedure and current prediction algorithms of the sales forecasting, then puts forward a prediction frame based on data mining, collaborates the decision-making tree, Bayesian Network model and association rules, and finally simulates it in the prediction of the sales and profits of the first quarter. Experimental results show that the predicted data well agrees with the actual data with bright research value.