基于决策树的汽车配置规则预测系统研究

Automatic Learning of Vehicle Configuration Rules Based on a Decision Tree Model

个性化汽车配置在欧美国家是一种流行的购车方案,目前配置规则判断系统研究领域尚属空白。本文提出了用于自动且有效率地判断客户特定需求的汽车订单是否有效的决策树模型,通过使用SQLServer提供的Analysis Service中的决策树算法和Weka的J48(C4.5算法)分别生成的决策树模型,对不同质量的配置订单群中训练数据和测试数据有效性进行了对比。本文在基于数据挖掘和决策树的研究基础之上,提出了基于决策树的汽车配置规则的预测系统,并且提出了软件系统的设计方案。实验结果表明该配置规则判断系统具有较好的实际应用价值。

Specific vehicle configuration is a popular vehicle purchase method in the countries of Europe and The United States, while research on configuration rule-decision system remains blank. Decision tree, the most widely used model in data mining, is a powerful tool in classification and prediction study. The current problem to be solved is how to automatically and efficiently acquire vehicle configuration rules that specify necessary and sufficient conditions for a vehicle ordered by customers. In this paper, an automatic learning of vehicle configuration rules based on a decision tree model has been proposed based on data mining knowledge and decision tree. At the same time, the software solution is also provided here. Experimental results show that the proposed vehicle configuration rule-decision system presented good practical application value.