SIA OpenIR  > 光电信息技术研究室
Alternative TitleThe Research and Application of Business Intelligence in the Unionpay’s Business Analysis System
Thesis Advisor李一凡
Keyword数据仓库 Etl 联机分析处理 数据挖掘
Call NumberTP311.13/W34/2007
Degree Discipline模式识别与智能系统
Degree Name硕士
Degree Grantor中国科学院沈阳自动化研究所
Place of Conferral沈阳
Abstract随着国内金融行业的逐步开放,中国银联也面临着跨国银行卡组织的激烈竞争,跨国银行卡组织拥有先进的IT技术和经营管理经验,和中国银联相比,具有很大竞争优势。银联为了积极面对跨国银行卡组织的挑战,成为一个知名品牌,必须加快信息化建设,转变经营观念,从传统的以业务为中心转移到以客户为中心,而商业智能技术——数据仓库和数据挖掘正是银联信息化建设的重要保障。 本文首先分析了银联的实际业务需求,结合银联的具体业务特点,设计与实现了银联的数据仓库系统,着重对数据仓库技术在银联中的应用现状进行了详细表述;该系统采用总线式的设计架构,有很好的一致性和可扩展性;系统采用维度建模方法进行数据仓库的逻辑设计,维度建模方法能很好地提高系统查询性能,在逻辑设计基础上本文也进行了数据仓库的物理设计。同时本文也详细介绍了数据仓库的重点部分——ETL系统的设计和实现,该ETL系统采用模块化的设计方法,采用元数据作为驱动方式,加强了调度管理和监控的功能,使该ETL工具更具智能性和更好的适应性。 本文在完成银联数据仓库系统建设的基础上,详细分析了银联业务系统要实现的OLAP分析目标,介绍了数据挖掘技术在银联客户分类中的应用,首次尝试在银联数据仓库系统中构建客户分类模型。在客户分类模型的构建中,我们首先采用聚类技术进行客户群分类,然后使用改进的SLIQ算法构建分类模型,本文对SLIQ算法中的符号型属性处理方法及其剪枝算法进行改进,并对结果进行对比分析,得到了一个较为合理的客户分类模型,取得了很好的应用效果,从而为银联数据仓库系统开发应用提供了可借鉴的操作思路。
Other AbstractWith the domestic banking completely opening up, China unionpay also faces tough competition. Foreign bank card organizations have advanced IT technology and management experiences, with competitive advantage over China’s financial industry. To become a brand enterprise, China unionpay must accelerate the development of the information industry and change the management concept from the traditional business-centered shifted to a client-centered. And business intelligence technology—data warehouse and data mining is an important guarantee for building the information system in unionpay. This paper analyzes the actual business requirements of unionpay, and puts emphases on analysis of the application status of data warehouse in unionpay, then designs and implements the unionpay’s data warehouse system in light of the specific business characteristics. The system bases on the bus framework and has a good consistency and scalability. Its logic model uses the dimension modeling which can improve query performance. Then we design the physical model. We design and implement the key part of data warehouse—ETL. The ETL system uses model designing and meta-data-driven approach, strengthens the scheduling management and monitoring functions, which has more intelligent and better adaptability. Based on the data warehouse system, we analyze the OLAP goals and the data mining technology used in the customer classification system. We are the first one to try building the customer classification model, and we use clustering algorithm to classify the customers, then use SLIQ algorithm to build the classification model, and improve the process of category attributes and SLIQ’s pruning algorithm. The customer classification model is reasonable and has the good effect and provids valuable experiences of development and application of data warehouse system on China unionpay.
Contribution Rank1
Document Type学位论文
Recommended Citation
GB/T 7714
王磊. 商业智能在银联业务分析系统中的应用研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2007.
Files in This Item:
File Name/Size DocType Version Access License
10001_20042801472803(1985KB) 开放获取--Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[王磊]'s Articles
Baidu academic
Similar articles in Baidu academic
[王磊]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[王磊]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.