摘要
信用卡的反欺诈是金融机构风控行业中的关键部分。在过去,信用卡的欺诈检测是通过实体和关系建模实现的,但是在处理非常复杂的实体和关系时,传统关系型数据库还存在一些不足。随着大数据和人工智能的发展,检测信用卡交易的方法也在不断发展。笔者提出了基于Neo4j图谱检测信用卡欺诈的方法,将数据集的特征通过图数据库直观展示出来,并且通过FICO评分标准建立了FICO模型。实验结果表明,该方法可以显著提高信用卡欺诈的认定率。
Credit card anti fraud is a key part of financial institutions'wind control industry.In the past,fraud detection of credit cards was realized through entity and relationship modeling,but there are still some shortcomings in traditional relational database when dealing with very complex entities and relationships.With the development of big data and artificial intelligence,the method of detecting credit card transactions is also developing.The author proposes a method to detect credit card fraud based on Neo4j atlas.The characteristics of the data set are displayed intuitively through the graph database,and the FICO model is established through the FICO scoring standard.The experimental results show that this method can significantly improve the credit card fraud rate.
作者
张芸芸
方勇
黄诚
Zhang Yunyun;Fang Yong;Huang Cheng(College of Electronics and Information Engineering,Sichuan University,Chengdu Sichuan 610065,China;College of Cybersecurity,Sichuan University,Chengdu Sichuan 610065,China)
出处
《信息与电脑》
2018年第21期23-25,共3页
Information & Computer
