基于关联规则的购物卡欺诈检测研究

关联规则分析被认为是数据挖掘中最有效的研究模型,能够发现相关项目之间潜在有用的关联规则,从而为决策者提供决策支持或为政策法规的制定提供依据。零售业的竞争越来越激烈,关联规则被广泛地应用到零售行业的数据分析中,基于此,以购物卡为例,为了检测和预防购物卡欺诈,从事务购物卡数据库中抽取知识,分析购物卡欺诈的一般特性,以便得出正常的行为模式,对于零售业业务风险管理的提升有所帮助。

Research on Froud Detection of Credit Card Based on Association Rules

Association rules analysis are considered to be the best studied models for data mining. We try to find the potential and useful association rules existed between the related items in order to provide with decision-making support or to provide evidence for policy-making. With the fierce competition in retail, association rules are widely applied to the data analysis of retail. Let＇s take credit card as an example. For detecting and preventing from credit card fraud, we extract knowledge from transaction credit card databases and analyze the general characteristics of credit card fraud, so that we can obtain the ordinary active model. This is helpful for raising the fraud management of retail industry.