基于数据挖掘的商业银行CRM系统研究与设计

发布时间：2018-10-23 17:29

【摘要】：随着科技的进步、互联网的发展，CRM已经从管理理论体系逐渐转化为全新的商业理念。相关统计显示，全球CRM系统软件市场在2012年到2016年之间的年复合增长率预计达到9.09%。由此可见，企业对CRM系统的需求在不断增长。在各行各业中，最早实施CRM系统的是金融业。随着各种高端的信息技术被运用到CRM系统中，例如互联网、人工智能、数据挖掘和数据仓库等，金融行业凭借自身强大的技术支持，在CRM系统实施的过程中获得了显著成效。商业银行是金融行业中最具有代表性的一支，其客户之多更加显示出对CRM系统的需求。随着银行数据的日积月累，数据仓库逐渐取代了数据库的地位，成为数据挖掘的主要信息载体被运用到CRM系统中。在商业银行，数据增长的速度令人叹为观止，如何能够有效的从海量数据中发现潜在的客户信息成为当今商业银行关注的焦点。本文针对上述背景进行商业银行CRM系统设计。文中首先对国内外商业银行实施CRM系统的现状进行研究，，而后针对系统设计涉及到的理论进行研究。文中针对银行客户分类算法从两方面进行研究。一方面，本文提出一种基于决策树的银行客户分类方法，对原先的方法与新提出的方法进行比较，发现新提出的方法在效率上占有一定的优势。另一方面，本文提出基于聚类的银行客户聚类算法，通过模糊聚类的方式对小样本客户进行分类，然后运用基于决策树的银行客户分类算法对大样本客户进行全面的分类，并能够确定分类原因。最后，通过对商业银行CRM系统的需求分析，本文基于客户统一视图理论设计了商业银行CRM系统的逻辑架构、数据仓库，并对CRM系统的主要业务模块进行设计，将数据挖掘技术应用到商业银行CRM系统中。
[Abstract]:With the progress of science and technology and the development of Internet, CRM has been transformed from management theory system to new business idea. The global CRM software market is expected to grow at a compound annual rate of 9.09% between 2012 and 2016, according to statistics. Thus, the demand for CRM system is growing. In a variety of industries, the earliest implementation of the CRM system is the financial industry. With the application of various high-end information technologies to CRM systems, such as the Internet, artificial intelligence, data mining and data warehouse, the financial industry, with its own strong technical support, has achieved remarkable results in the implementation of the CRM system. Commercial bank is the most representative branch in the financial industry, and the number of customers shows the demand for CRM system. With the accumulation of bank data, data warehouse has gradually replaced the status of database and become the main information carrier of data mining in CRM system. In commercial banks, the speed of data growth is amazing, how to effectively find potential customer information from massive data has become the focus of commercial banks. This article carries on the commercial bank CRM system design according to the above background. In this paper, the current situation of commercial banks implementing CRM system at home and abroad is studied, and then the theory involved in the system design is studied. In this paper, the bank customer classification algorithm is studied from two aspects. On the one hand, this paper proposes a bank customer classification method based on decision tree, compares the original method with the new one, and finds that the new method has some advantages in efficiency. On the other hand, this paper proposes a clustering algorithm for bank customers, which classifies small sample customers by fuzzy clustering, and then makes a comprehensive classification of large sample customers by using the decision tree based bank customer classification algorithm. And can determine the reasons for classification. Finally, through the requirement analysis of commercial bank CRM system, this paper designs the logic structure, data warehouse and the main business module of CRM system based on the customer unified view theory. The data mining technology is applied to the commercial bank CRM system.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP311.13;F832.2

【参考文献】