汽车行业销售信息分析系统的分析与设计
摘要
随着信息技术的深入应用,企业所面临的市场竞争日益加剧,面对的市场环境日趋复杂。如何利用好企业所掌握的数据信息对市场的变化做出快速的反应,成为企业赢得市场的关键。企业的经营管理者希望能够对市场的变化迅速地做出科学合理的决策。而大部分企业现有的各种业务信息系统很难对中高层的决策支持起到作用,满足不了目前信息处理的要求。企业的大量历史数据的不到有效的利用,造成数据的浪费而不能支持企业决策。信息分析系统和数据仓库为企业提供了处理所需要的决策信息的有效方法,因此,对于企业来说很有必要建设信息分析系统和合理构建数据仓库,并以此为基础进行数据的分析处理和数据挖掘从而为经营管理者提供科学的决策依据,高效的制订决策方案。因而信息分析系统建设的研究也越来越受到人们的重视。
本文以对某汽车企业信息分析系统开发的实际项目为基础,对企业信息分析系统及数据仓库建设方面作了较多的研究和尝试。通过对企业信息分析系统发展的过程、现有理论的学习和研究,实际完成企业信息分析系统设计与实现任务。在对企业实际需求进行分析的基础上,利用J2EE技术及数据挖掘技术对该信息分析系统进行设计与实施。文中介绍了基于J2EE技术的系统三层架构模式以及数据仓库系统的开发模式,对建立数据模型、数据存储以及数据的提取、转换和加载进行了说明,详细阐述了信息分析系统的设计与实现过程。
The 21st century is the era of the Internet economy, information technology and e-commerce was a new development. With the application of information technology, enterprises are confronted with more and more acute market competition and the market environment become more and more complicated. To enterprises, information and business activities in the increasingly important status. Intense competition requires the enterprises to take advantage of new technologies on information management and analysis. So how to utilize the data information and make a quickly reaction to the various market is vital for the enterprises to success in these competition. Enterprise administrators hope to make a scientific and reasonable decision in the changeable market. However, nowadays the business information systems in enterprise cannot support the decision for middle and upper levels and also cannot satisfy current information processing. A great lot of history data cannot be used for the enterprise. Many useful data wasted and cannot support the decision for the enterprise. As information analysis system and data warehouse (DW) is an effective processing tool for the enterprise, it is necessary for the enterprise to construct the information analysis system and data warehouse that can be linked on it to make the disposal of data analysis and data mining. With it to get a scientific reference for the enterprise management decision and to get a more highly effective to make blue print for the enterprise. Therefore the research of construction information analysis system has been becoming more and more important.
This article based on the construction of information analysis system project for one automobile corporation, and has done more research and attempt on enterprise information analysis system and data warehouse construction. Finally,finished the design and construction through studying the information analysis system development process and existing theory. In this project the technical of J2EE and data mining were used in the foundation of analysis the enterprise actual demand to design the information analysis system. Papers detail of the information analysis system design, including the design of data warehouse and analysis system designed. And the basis of a specific module to complete the implementation of the program, the Data Analysis module which is the sales analysis system to the core , Above the basis of Data Mining Techniques, better use of data mining methods in the mainstream of technology --J2EE B/S development of the information technology analysis system Development . That the establishment of a strong auto analysis module targeted for enterprise decision-makers provided an effective basis for decision-making .The article introduced the three–layer employer model based on technical of J2EE and DW system development pattern, and explained the establishment of data model, data storage as well as data extracting and transforming and loading, elaborated the information analysis system design and realization process.
The system design and implementation has practical meaning, throughout the system design and development process the theoretical knowledge was applied to practical work. In this project, the practical application of experience was acquired.
Meantime we have encountered some actual problem in our work. Those issues will be having further studied. The implementation of the project on the auto corporation will improve its information technology and information analysis capabilities, we also make some conclusions:
1. A substantial increase in the historical data and daily data to be used efficiency, a result of these sales data analysis, Decision-makers could to make better analysis of the product sales during previous time and to use the actual sales data to support decision-making.
2. Transform a great deal of selling data processing from manual processing into computer automatic processing. Operational staff release from a lot of manual processing work.
3. The use of mainstream technology for the development of this system , the use of modular design make this system has a better stability, high speed feedback and high security advantages ,improve the system's efficiency.
4. Integrate the data mining and data warehouse technology, and combine system development with our subject knowledge, and those were implemented in practical work and the system design.
5. The implementation of System Framework Technology, which makes the whole system a better scalability. System has good scalability and good maintainability. In the future, system expansion only need to add the corresponding functional modules can be achieved system maintenance, upgrading.
6. Upgrade the auto corporation’s information construction and information analysis capability. It has good practical significance in the future information construction.
本文以对某汽车企业信息分析系统开发的实际项目为基础,对企业信息分析系统及数据仓库建设方面作了较多的研究和尝试。通过对企业信息分析系统发展的过程、现有理论的学习和研究,实际完成企业信息分析系统设计与实现任务。在对企业实际需求进行分析的基础上,利用J2EE技术及数据挖掘技术对该信息分析系统进行设计与实施。文中介绍了基于J2EE技术的系统三层架构模式以及数据仓库系统的开发模式,对建立数据模型、数据存储以及数据的提取、转换和加载进行了说明,详细阐述了信息分析系统的设计与实现过程。
The 21st century is the era of the Internet economy, information technology and e-commerce was a new development. With the application of information technology, enterprises are confronted with more and more acute market competition and the market environment become more and more complicated. To enterprises, information and business activities in the increasingly important status. Intense competition requires the enterprises to take advantage of new technologies on information management and analysis. So how to utilize the data information and make a quickly reaction to the various market is vital for the enterprises to success in these competition. Enterprise administrators hope to make a scientific and reasonable decision in the changeable market. However, nowadays the business information systems in enterprise cannot support the decision for middle and upper levels and also cannot satisfy current information processing. A great lot of history data cannot be used for the enterprise. Many useful data wasted and cannot support the decision for the enterprise. As information analysis system and data warehouse (DW) is an effective processing tool for the enterprise, it is necessary for the enterprise to construct the information analysis system and data warehouse that can be linked on it to make the disposal of data analysis and data mining. With it to get a scientific reference for the enterprise management decision and to get a more highly effective to make blue print for the enterprise. Therefore the research of construction information analysis system has been becoming more and more important.
This article based on the construction of information analysis system project for one automobile corporation, and has done more research and attempt on enterprise information analysis system and data warehouse construction. Finally,finished the design and construction through studying the information analysis system development process and existing theory. In this project the technical of J2EE and data mining were used in the foundation of analysis the enterprise actual demand to design the information analysis system. Papers detail of the information analysis system design, including the design of data warehouse and analysis system designed. And the basis of a specific module to complete the implementation of the program, the Data Analysis module which is the sales analysis system to the core , Above the basis of Data Mining Techniques, better use of data mining methods in the mainstream of technology --J2EE B/S development of the information technology analysis system Development . That the establishment of a strong auto analysis module targeted for enterprise decision-makers provided an effective basis for decision-making .The article introduced the three–layer employer model based on technical of J2EE and DW system development pattern, and explained the establishment of data model, data storage as well as data extracting and transforming and loading, elaborated the information analysis system design and realization process.
The system design and implementation has practical meaning, throughout the system design and development process the theoretical knowledge was applied to practical work. In this project, the practical application of experience was acquired.
Meantime we have encountered some actual problem in our work. Those issues will be having further studied. The implementation of the project on the auto corporation will improve its information technology and information analysis capabilities, we also make some conclusions:
1. A substantial increase in the historical data and daily data to be used efficiency, a result of these sales data analysis, Decision-makers could to make better analysis of the product sales during previous time and to use the actual sales data to support decision-making.
2. Transform a great deal of selling data processing from manual processing into computer automatic processing. Operational staff release from a lot of manual processing work.
3. The use of mainstream technology for the development of this system , the use of modular design make this system has a better stability, high speed feedback and high security advantages ,improve the system's efficiency.
4. Integrate the data mining and data warehouse technology, and combine system development with our subject knowledge, and those were implemented in practical work and the system design.
5. The implementation of System Framework Technology, which makes the whole system a better scalability. System has good scalability and good maintainability. In the future, system expansion only need to add the corresponding functional modules can be achieved system maintenance, upgrading.
6. Upgrade the auto corporation’s information construction and information analysis capability. It has good practical significance in the future information construction.
引文
[1]池太葳,《数据仓库结构设计与实施——建造信息系统的金字塔》,电子工业出版社,2005 年 11 月
[2][美]Ralph Kimball、 Margy Ross、谭明金 (译),《数据仓库工具箱——维度建模的完全指南》(第二版),电子工业出版社, 2003 年 10 月
[3][美]Alex Berson、Stephen Smith、Kurt Thearling, 《构建面向 CRM的数据挖掘应用》,人民邮电出版社, 2001 年 8 月
[4][美]Tom Soukup、 Ian Davidson 、朱建秋、蔡伟杰 (译) ,《可视化数据挖掘——数据可视化和数据挖掘的技术与工具》, 电子工业出版社, 2004年 1 月
[5][美]Ryszard S、Michalski、 Ivan Bratko、 Miroslav Kubat 等、朱明等(译), 《机器学习与数据挖掘——方法和应用》, 电子工业出版社, 2004年 1 月
[6][美]Gordon S.Linoff 、Michael J.A.Berry, 《Web 数据挖掘》, 电子工业出版社,2004 年 3 月
[7]郑阿齐、刘启芬、 顾韵华 , 《SQL Server 实用教程》(第二版), 电子工业出版社,2005 年 7 月
[8]李子奈, 《计量经济学》, 高等教育出版社, 1998 年
[9]陶树平、屠颖,《关联规则和分类规则挖掘算法的改进与实现》,计算机工程 2003 年 9 月
[10] V.Fgurbaxani,《The Impact of Information Systems on Organizations and Markets》, Communication of the ACh1, 1991,Vol.34 (l):59-73
[11]Ivar Jacobson,《Software Reuse Architecture Process and Organization for Business Success》,世界图书出版公司,NewYork,1998年
[12] 杨学兵、刘胜军等,《一种实时过程控制中的数据挖掘算法研究》,计算机应用,1999 年第 9 期第 19 卷
[13] R. Binder, 《Testing Object-Oriented System: A Status Report》, American Programmer,1994
[14]R. Otte、 P. Patrick、 M. Roy、李师贤等译,《CORBA 教程:公共对象请求代理体系结构》,清华大学出版社,1999 年
[15]Richard J.Roiger,Michael W.Geatz 、翁敬弄 (译),《Data Mining A Tutorial-Based Primer》,清华大学出版社,2003 年
[16]孟祥萍、钱进、刘大有,《基于数组的关联规则挖掘算法》,计算机工程,2003 年 9 月
[17]李德毅,《一种测试数据挖掘算法的数据源生成方法》,计算机研究与发展,2000 年第 37 卷第 7 期
[18]Margaret.H.Dunham,《DATA MING Introductory and Advanced Topics》,清华大学出版社,2003 年
[19]Ralph Kimball、Margy ross、谭明金(译),《The Data Warehouse Toollcitahe Complete Guide to Dimensional Modeling》,电子工业出版社,2003 年 10 月
[20]Micheal Corey 、Micheal Abbey,《SQLSERVER 7 Data warehousing》,希望电子出版社,2000 年
[21]陈文伟、黄金才、赵新显,《数据挖掘技术》,北京工业大学出版社,2002 年 12 月
[22]Tony Bain、邵勇,《SQL Server 2000 数据仓库与 Analysis Services》,中国电力出版社,2002 年 11 月
[23] 李勇、肖智、陈玲,《汽车销售分析与预测系统的设计与开发》,重庆人学学报第 25 卷第 9 期,2002 年 9 月
[23]陈文伟,《决策支持系统及其开发》(第二版),清华大学出版社,2000年
[24]余庆、朱玉全,《基于 FP-Tree 的最大频繁项目集挖掘及更新算法》,软件学报,2003 年 14 卷
[25]范明、李川,《在 FP2 树中挖掘频繁模式而不生成条件 FP2 树》,计算机研究与发展,2003 年 8 月第 40 卷第 8 期
[26]朱玉全、孙志挥、赵传申,《快速更新频繁项集》,计算机研究与发展,2003 年 1 月第 40 卷
[27]尹旭日、陈世福,《一种基于 Rough 集的缺省规则挖掘算法》,计算机研究与发展,2000 年 12 月第 37 卷
[28]李绪成、王保保,《挖掘关联规则中 Apriori 算法的一种改进》,计算机工程,2002 年 7 月第 28 卷 104-105
[29]颜雪松、蔡之华,《一种基于 Apriori 的高效关联规则挖掘算法的研究》,计算机工程与应用,2002 年 10 月
[30]冯玉才、冯剑琳,《关联规则的增量式更新算法》,软件学报,1998 年第 9 期
[31]范明、孟小峰等译,《数据挖掘概念与技术》,机械工业出版社,2001年
[32]张华,《基于大数据集聚类分析研究》(硕士学位论文),华东理工大学,2002 年
[33]张云涛、龚玲,《数据挖掘原理与技术》,电子工业出版社,2004 年
[34]罗积玉、刑瑛,《经济统计分析方法及预测〉,清华大学出版社,1987年
[35]李仲来,《系统聚类公式的推广》,应用概率统计,1994 年 10 月
[36]吴元奇、冯容杨,《聚类分析计算方法的理论及结果比较》,湛江海洋大学学报,2002 年第 22 期 [37」陈文伟、黄金才、陈元,《决策支持系统新式结构体系》,决策与决策支持系统,1998 年
[38]王守惠、张全寿,《决策支持系统研究的发展趋势及前景》,决策与决策支持系,1996 年 6 月
[40]朱建秋,《数据挖掘系统发展综述》,上海复旦德门软件有限公司,2004年
[41]Richard 、Kirkby、Eibe Frank,《WEKA Explorer User Guide for Version》,2004 年三月
[42]周健吉,《数据挖掘工具的评判》,电脑知识与技术,2004 年 23 卷
[43]蔡伟杰,《可视化数据挖掘技术研究及实现》,计算机应用,2001 年 4月
[44]陈文伟,《决策支持系统及其开发》(第 3 版),清华大学出版社,2003年
[45]段晓君等,《可视化数据挖掘技术及其应用》,电脑开发与应用,2000年 1 月
[46]王斌会等,《数据挖掘技术及其应用现状》,软件学报,2004 年 2 月
[47]赵铁军,《文本信息结构化与可视化研究》,计算机应用研究,2004 年7 月
[48]程岩、黄梯百,《信息系统中一种面向粗糙集的数据挖掘方法》,情报学报,2001 第 20 卷
[49]汤字松、刘相峰等,《数据挖掘系统设计系统》,工程理论实践,2000年第 9 期
[50]周腾,《数据挖掘、数据仓库构架的拓展》,电脑与信息技术,2001 第2 期
[51]王燕,《数据挖掘技术应用研究》,甘肃科技信息技术,2001 第 49 期
[52]赵丹群,《数据挖掘原理、方法及其应用》,现代图书情报技术,2000年第 6 期
[53]张红百、马垣,《数据挖掘与决策支持系统的关系》,鞍山师范学院学报,2001 第 9 期
[2][美]Ralph Kimball、 Margy Ross、谭明金 (译),《数据仓库工具箱——维度建模的完全指南》(第二版),电子工业出版社, 2003 年 10 月
[3][美]Alex Berson、Stephen Smith、Kurt Thearling, 《构建面向 CRM的数据挖掘应用》,人民邮电出版社, 2001 年 8 月
[4][美]Tom Soukup、 Ian Davidson 、朱建秋、蔡伟杰 (译) ,《可视化数据挖掘——数据可视化和数据挖掘的技术与工具》, 电子工业出版社, 2004年 1 月
[5][美]Ryszard S、Michalski、 Ivan Bratko、 Miroslav Kubat 等、朱明等(译), 《机器学习与数据挖掘——方法和应用》, 电子工业出版社, 2004年 1 月
[6][美]Gordon S.Linoff 、Michael J.A.Berry, 《Web 数据挖掘》, 电子工业出版社,2004 年 3 月
[7]郑阿齐、刘启芬、 顾韵华 , 《SQL Server 实用教程》(第二版), 电子工业出版社,2005 年 7 月
[8]李子奈, 《计量经济学》, 高等教育出版社, 1998 年
[9]陶树平、屠颖,《关联规则和分类规则挖掘算法的改进与实现》,计算机工程 2003 年 9 月
[10] V.Fgurbaxani,《The Impact of Information Systems on Organizations and Markets》, Communication of the ACh1, 1991,Vol.34 (l):59-73
[11]Ivar Jacobson,《Software Reuse Architecture Process and Organization for Business Success》,世界图书出版公司,NewYork,1998年
[12] 杨学兵、刘胜军等,《一种实时过程控制中的数据挖掘算法研究》,计算机应用,1999 年第 9 期第 19 卷
[13] R. Binder, 《Testing Object-Oriented System: A Status Report》, American Programmer,1994
[14]R. Otte、 P. Patrick、 M. Roy、李师贤等译,《CORBA 教程:公共对象请求代理体系结构》,清华大学出版社,1999 年
[15]Richard J.Roiger,Michael W.Geatz 、翁敬弄 (译),《Data Mining A Tutorial-Based Primer》,清华大学出版社,2003 年
[16]孟祥萍、钱进、刘大有,《基于数组的关联规则挖掘算法》,计算机工程,2003 年 9 月
[17]李德毅,《一种测试数据挖掘算法的数据源生成方法》,计算机研究与发展,2000 年第 37 卷第 7 期
[18]Margaret.H.Dunham,《DATA MING Introductory and Advanced Topics》,清华大学出版社,2003 年
[19]Ralph Kimball、Margy ross、谭明金(译),《The Data Warehouse Toollcitahe Complete Guide to Dimensional Modeling》,电子工业出版社,2003 年 10 月
[20]Micheal Corey 、Micheal Abbey,《SQLSERVER 7 Data warehousing》,希望电子出版社,2000 年
[21]陈文伟、黄金才、赵新显,《数据挖掘技术》,北京工业大学出版社,2002 年 12 月
[22]Tony Bain、邵勇,《SQL Server 2000 数据仓库与 Analysis Services》,中国电力出版社,2002 年 11 月
[23] 李勇、肖智、陈玲,《汽车销售分析与预测系统的设计与开发》,重庆人学学报第 25 卷第 9 期,2002 年 9 月
[23]陈文伟,《决策支持系统及其开发》(第二版),清华大学出版社,2000年
[24]余庆、朱玉全,《基于 FP-Tree 的最大频繁项目集挖掘及更新算法》,软件学报,2003 年 14 卷
[25]范明、李川,《在 FP2 树中挖掘频繁模式而不生成条件 FP2 树》,计算机研究与发展,2003 年 8 月第 40 卷第 8 期
[26]朱玉全、孙志挥、赵传申,《快速更新频繁项集》,计算机研究与发展,2003 年 1 月第 40 卷
[27]尹旭日、陈世福,《一种基于 Rough 集的缺省规则挖掘算法》,计算机研究与发展,2000 年 12 月第 37 卷
[28]李绪成、王保保,《挖掘关联规则中 Apriori 算法的一种改进》,计算机工程,2002 年 7 月第 28 卷 104-105
[29]颜雪松、蔡之华,《一种基于 Apriori 的高效关联规则挖掘算法的研究》,计算机工程与应用,2002 年 10 月
[30]冯玉才、冯剑琳,《关联规则的增量式更新算法》,软件学报,1998 年第 9 期
[31]范明、孟小峰等译,《数据挖掘概念与技术》,机械工业出版社,2001年
[32]张华,《基于大数据集聚类分析研究》(硕士学位论文),华东理工大学,2002 年
[33]张云涛、龚玲,《数据挖掘原理与技术》,电子工业出版社,2004 年
[34]罗积玉、刑瑛,《经济统计分析方法及预测〉,清华大学出版社,1987年
[35]李仲来,《系统聚类公式的推广》,应用概率统计,1994 年 10 月
[36]吴元奇、冯容杨,《聚类分析计算方法的理论及结果比较》,湛江海洋大学学报,2002 年第 22 期 [37」陈文伟、黄金才、陈元,《决策支持系统新式结构体系》,决策与决策支持系统,1998 年
[38]王守惠、张全寿,《决策支持系统研究的发展趋势及前景》,决策与决策支持系,1996 年 6 月
[40]朱建秋,《数据挖掘系统发展综述》,上海复旦德门软件有限公司,2004年
[41]Richard 、Kirkby、Eibe Frank,《WEKA Explorer User Guide for Version》,2004 年三月
[42]周健吉,《数据挖掘工具的评判》,电脑知识与技术,2004 年 23 卷
[43]蔡伟杰,《可视化数据挖掘技术研究及实现》,计算机应用,2001 年 4月
[44]陈文伟,《决策支持系统及其开发》(第 3 版),清华大学出版社,2003年
[45]段晓君等,《可视化数据挖掘技术及其应用》,电脑开发与应用,2000年 1 月
[46]王斌会等,《数据挖掘技术及其应用现状》,软件学报,2004 年 2 月
[47]赵铁军,《文本信息结构化与可视化研究》,计算机应用研究,2004 年7 月
[48]程岩、黄梯百,《信息系统中一种面向粗糙集的数据挖掘方法》,情报学报,2001 第 20 卷
[49]汤字松、刘相峰等,《数据挖掘系统设计系统》,工程理论实践,2000年第 9 期
[50]周腾,《数据挖掘、数据仓库构架的拓展》,电脑与信息技术,2001 第2 期
[51]王燕,《数据挖掘技术应用研究》,甘肃科技信息技术,2001 第 49 期
[52]赵丹群,《数据挖掘原理、方法及其应用》,现代图书情报技术,2000年第 6 期
[53]张红百、马垣,《数据挖掘与决策支持系统的关系》,鞍山师范学院学报,2001 第 9 期