动态环境下多个时期的客户购物模式变化挖掘  

Change mining of customer shopping patterns from multi-period datasets under dynamic environment

作　　者：朱慧云[1] 陈森发[1] 张丽杰[2] 

机构地区：[1]东南大学经济管理学院,南京210096 [2]南京信息工程大学经济管理学院,南京210044

出　　处：《东南大学学报（自然科学版）》2012年第5期1012-1015,共4页Journal of Southeast University：Natural Science Edition

基　　金：教育部博士点基金资助项目(2006028600);江苏省高校自然科学研究资助项目(10KJB170011)

摘　　要：针对现有多个时期变化挖掘方法的不足,在已有的稳定和趋势变化类型基础上,提出了突变规则的概念,将支持度(或置信度)序列中存在孤立点的规则定义为突变规则,并使用格拉布斯检验发现突变规则.突变规则表明客户购物行为的突变,可以更全面地反映客户购物模式随时间变化的动态信息,帮助管理者根据变化模式及时采取有效的应对措施,取得竞争优势.为了对变化模式的重要程度进行评估,将规则的支持度(或置信度)均值作为稳定规则兴趣度的客观度量,将Mann-Kendall检验的Kendall斜率作为趋势变化的显著程度度量,并将孤立点的偏离程度作为突变规则兴趣度的度量.实证分析结果表明,所提方法可以有效地识别多个时期数据集的客户购物模式变化.Aiming at the shortcomings of existing change mining methods from multi-period data- sets, the concept of abrupt rules based on two existing types of changes, stable rules and those which exhibit trends, is presented. The rules that outliers exist in support or confidence time series are de- fined as abrupt rules, and the Grubbs test is used to discover the abrupt rules. The abrupt rules ex- hibit the abrupt changes of customer shopping behavior, which can reflect the dynamic information of customer shopping patterns with time comprehensively and help managers take effective strategies timely to gain competitive advantage. In order to evaluate the degree of change, the average value of support or confidence of rules is used as the interest measure for stable rules; the Kendall slope of the Mann-Kendall test is used as the degree of change for rules which exhibit trends; and the departure degree of outliers is used as the interest measure for abrupt rules. The results of empirical evaluation show that the proposed method can effectively recognize the change of customer shopping patterns from multi-period datasets.

关 键 词：数据挖掘 变化分析 变化挖掘 关联规则 零售业 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]