反欺诈模型在电子银行的应用研究

发布时间：2018-03-14 06:04

本文选题：电子银行　切入点：反欺诈　出处：《山东大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着商业银行电子银行市场规模的逐步扩大,针对电子银行的欺诈行为也与之俱来,如何在巨量的客户交易与可疑行为中有效识别和防范欺诈,成为众多商业银行研究的内容和目标。本文努力找到并搭建一个切实可行、可操作、投入与产出相匹配的模型,希望能对我国中小银行探索电子银行反欺诈工作提供思路和帮助。本文首先从当前电子银行市场上主流的欺诈行为分析入手,阐述当前困扰商业银行的欺诈行为的主要欺诈方式和欺诈特点,让读者对目前的电子银行欺诈行为有所了解和认识,有针对性地引出本文重点研究的对象—电子银行反欺诈。其次,论文从两方面论证分析了商业银行开展电子银行反欺诈行为的紧迫性与重要意义,结合目前国内主流的反欺诈模型理论,提出了商业银行建设自身的反欺诈系统时、在模型的理论应用时需要重点关注和解决的问题。接下来的内容是本文的重点,即是在对欺诈行为分析的基础上,详实、详尽地介绍了国内现有的两种电子银行反欺诈模型：基于规则的反欺诈模型与基于客户行为的反欺诈模型。针对两种模型的工作原理、运作机制作了详尽的阐述,并以案例的形式解释模型的实际应用。两种模型的理论基础都是基于对电子银行客户行为或交易的风险暴露,其本质是按照既定的规则引擎和分析标准对巨量客户行为和事务数据的筛选。在对两种模型分析和研究的基础上,总结性的针对两种模型的优点和劣势进行了论证和分析。本文通过全面、完整的论述两种模型的运作机理后,也重点将模型的实际应用情况作了分析和论证,并提出了“两种模型交叉使用”作为本文在模型实际应用方面的观点。为支持本文的观点,也引用了模型应用的成功案例,并就模型应用过程中仍然存在的疑点、难点、障碍等作了总结与分析,旨在说明商业银行在反欺诈模型应用时,将客户巨量行为数据分析、客户信息资源共享、巨量的欺诈行为与反欺诈行为的系统精准性调试作为模型的应用改进方向,使模型更好的发挥作用,帮助构建和夯实商业银行的反欺诈体系。最后本论文对研究的内容和结果进行了总结,并指出反欺诈模型的应用改进和优化的方向方法,总体来讲,具有以下几方面的创新： 1.对比研究了国内目前主流的两种电子银行反欺诈模型的概念、运作原理和优缺点,提出了一个适合当前我国中小银行搭建的电子银行反欺诈模型。 2.总结性地对当前制约反欺诈模型实际应用过程中的难点、缺点进行对比与分析,提出针对提升实际应用效果的模型应用改进。 3.创造性的提出反欺诈系统的运用需要借助金融系统大数据化的资源整合和信息共享,才能达到理想效果。
[Abstract]:With the gradual expansion of commercial bank electronic banking market scale, fraud against electronic banking has come along with it, how to effectively identify and prevent fraud in a large number of customer transactions and suspicious behavior. This paper tries to find and build a feasible, operable, input-output matching model. Hope to provide ideas and help for small and medium-sized banks to explore e-bank anti-fraud work. This paper begins with the analysis of the current mainstream fraud in the electronic banking market, and expounds the main fraud ways and the characteristics of the fraud that puzzles the commercial banks at present. Let readers have some understanding of the current electronic banking fraud behavior, and specifically lead to the focus of this study-electronic banking anti-fraud. Secondly, This paper analyzes the urgency and significance of commercial banks to carry out electronic banking anti-fraud behavior from two aspects. Combining with the current mainstream anti-fraud model theory, the paper puts forward the commercial banks to build their own anti-fraud system. In the theoretical application of the model, we need to pay attention to and solve the problems. The following content is the focus of this paper, that is, on the basis of the analysis of fraud, This paper introduces in detail two kinds of electronic bank anti-fraud models in China: rule-based anti-fraud model and anti-fraud model based on customer behavior. The theoretical basis of the two models is based on the risk exposure to the behavior or transaction of electronic bank customers. Its essence is to screen huge amount of customer behavior and transaction data according to the established rule engine and analysis standard. On the basis of analyzing and studying the two models, the paper demonstrates and analyzes the advantages and disadvantages of the two models. After a comprehensive and complete discussion on the operation mechanism of the two models, this paper also focuses on the analysis and demonstration of the practical application of the models. In order to support the viewpoint of this paper, the successful cases of model application are also cited, and the doubtful points and difficulties that still exist in the process of model application are discussed. The purpose of this paper is to analyze the huge amount of customer behavior data and share customer information resources in the application of anti-fraud model. Large amount of fraud and anti-fraud system precision debugging as the direction of improvement of the application of the model, make the model play a better role, help to build and consolidate the anti-fraud system of commercial banks. Finally, this paper summarizes the contents and results of the research, and points out the direction of application improvement and optimization of anti-fraud model, generally speaking, has the following aspects of innovation:. 1. The concept, operation principle, advantages and disadvantages of two kinds of electronic bank anti-fraud models are studied, and an electronic bank anti-fraud model which is suitable for the small and medium-sized banks in China is put forward. 2. Compare and analyze the difficulties and shortcomings in the practical application of the current anti-fraud model, and put forward the model application improvement aimed at improving the practical application effect. 3. It is creatively proposed that the application of anti-fraud system needs the integration of resources and information sharing of the financial system to achieve the desired results.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F724.6;F832.2;F224

【参考文献】