数据仓库和数据挖掘在商业银行客户关系管理中的应用

发布时间：2018-02-11 05:15

本文关键词： 数据仓库数据挖掘客户关系管理 OLAP K-Means X-Means　出处：《长安大学》2013年硕士论文　论文类型：学位论文

【摘要】：随着计算机技术的发展，数据仓库和数据挖掘技术也越来越多地被应用到社会的各个行业，尤其是那些存储着海量客户信息的行业：如电信业、金融业、零售业等。这些行业之间的竞争归根结底是对客户归属处的竞争，因此掌握客户信息，从众多客户信息中挖掘出对自己有利的信息则至关重要。数据仓库和数据挖掘技术可以帮助这些企业来分析客户信息，并且可以预测客户的消费趋势，帮助企业留住现有客户，挖掘潜在客户，从而帮助企业降低成本，增加收入，在竞争中立于不败之地。本文主要研究的是数据仓库和数据挖掘技术在银行客户关系管理中的应用。以银行客户信息为背景，利用数据仓库和数据挖掘技术，主要做了以下工作：（1）基于Solaris10的DB2数据库建立了数据仓库模型。模型有三个模式，ODS、EDW和DDW模式。模式和模式之间的数据导入是通过ETL来实现的。（2）利用SQLServer的BIDS组件的OLAP技术分析客户账户信息、客户贷款账户信息和银行各分行的交易信息。（3）利用数据挖掘技术，对数据仓库中的数据进行再处理。使用数据用例分析了DBSCAN算法、K-Means算法和X-Means算法的优缺点，最后采用X-Means算法来实现个人客户群体细分。在此基础上，利用FP-Growth算法实现了银行业务的交叉营销。本文分析内容有一定的实际和理论意义，，可以为银行的决策者提供科学的决策支持。
[Abstract]:With the development of computer technology, data warehouse and data mining technology are more and more applied to every industry of society, especially those which store huge amount of customer information, such as telecommunication industry, financial industry, Retailing and so on. The competition between these industries is ultimately a competition for customer ownership, so we have customer information. Data warehouse and data mining technology can help these enterprises to analyze customer information, and can predict customer consumption trend and help enterprises retain existing customers. Mining potential customers to help enterprises reduce costs, increase revenue, in the competition in an invincible position. This paper mainly studies the application of data warehouse and data mining technology in bank customer relationship management. With the background of bank customer information, using data warehouse and data mining technology, the main work is as follows:. 1) the data warehouse model is established based on the DB2 database of Solaris10. There are three modes in the model: ODS-EDW and DDW. The data import between schema and schema is realized by ETL. Using OLAP technology of BIDS component of SQLServer to analyze customer account information, customer loan account information and transaction information of bank branches. Data mining technology is used to reprocess the data in data warehouse. The advantages and disadvantages of DBSCAN algorithm, K-Means algorithm and X-Means algorithm are analyzed by using data case. Finally, X-Means algorithm is used to realize individual customer group segmentation. The cross-marketing of bank business is realized by FP-Growth algorithm. This paper has some practical and theoretical significance and can provide scientific decision support for bank decision-makers.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP311.13

【参考文献】