数据挖掘在零售企业信息化建设中的应用研究

VIP免费

3.0 赵德峰 2024-11-19 4 4 789.07KB 64 页 15积分

侵权投诉

摘要

近年来，随着科学技术的飞速发展，经济和社会都取得了极大的进步，与此

同时，在各个领域产生了大量的数据，激增的数据背后隐藏着许多重要的信息。

人们不再满足于数据库的查询功能，希望能够对其进行更高层次的分析，以便能

从数据中提取信息和知识为决策服务。数据挖掘的出现为自动和智能地把海量的

数据转化为有用的信息和知识提供了手段，它把人们对数据的应用从低层次的简

单查询，提升到从数据中挖掘知识，提供决策支持。

本文首先以某茶叶有限公司为依托，深入分析数据挖掘中的模糊聚类与关联

规则技术，重点讨论基于模糊相似关系聚类的传递闭包法、基于关系代数理论的

关联规则挖掘算法——ORAR、以及基于概念分层的泛化关联规则挖掘算法——

RGAR。其次，在理论分析的基础上，结合某茶叶有限公司的实际情况和迫切需

求，利用模糊聚类技术分析公司与超市买场客户之间的历史交易记录数据，并结

合客户价值理论，获得较为客观和科学的客户价值评价；利用关联规则分析不同

顾客购物篮数据，以获得顾客在购买某种产品的同时再购买其他产品的可能性，

这样使公司决策者在进行客户关系维护、产品分析以及库存管理等方面有据可

依。最后，在以上两部分的理论分析和实际应用的基础上，本文针对某茶叶有限

公司的实际情况，以软件包的形式开发基于模糊聚类、关联规则的决策支持系统，

从零售企业信息化的角度改善企业的管理模式与业务流程，达到提高企业管理水

平的最终目的。

关键词：数据挖掘模糊聚类关联规则决策支持

ABSTRACT

In recent years, with the rapid development of science and technology, both

economy and society have made remarkable progress. Meanwhile, different fields all

produce large quantities of data, which conceal plenty of important information and

people no longer feel content with the retrieval function of database, rather we hope to

do the analysis on higher level. So that information or knowledge extracted from the

data can aid our decision making process. Data mining can automatically and

intelligently transform large number of data into useful information and knowledge.

So data mining, as such an effective means, can actually provide decision support.

On the basis of actual conditions of Shanghai Qunfeng Tea Ltd. Co., thorough

analysis of fuzzy clustering and association rules, this paper mainly discusses

Transitive Closure Method on fuzzy clustering analysis, optimization relation

association rule on association algebra, and generalized association rule on the

concept stratification. Secondly, on the basis of theoretical analysis and actual

conditions of Shanghai Qunfeng Tea Ltd. Co.. The writer uses fuzzy clustering to

analyze the history business records data between the company and customers, and

tries to gain an objective and scientific customer value evaluation result by the

customer value theory. Besides, the writer uses association rule in analyzing the

buying data of different customers to know the possibility of buying additional

products when the customers are shopping. In this way, the decision-makers of the

company can rely on the above analysis to conduct customer relation maintenance,

product analysis and inventory management. Finally, on the basis of the above

theoretical analysis and practical application, the writer develops a decision support

system on fuzzy clustering and association rule, which is oriented to Shanghai

Qunfeng Tea Ltd. Co.. The system proves to be successful, because the management

pattern and business procedure of the company are improved from the

informationization perspectives of retailing enterprise, and its management level is

also raised in the end.

Key Word: data mining, fuzzy clustering, association rule, decision

support

摘要

ABSTRACT

第一章绪论 ···················································································· 1

§1.1 论文研究的背景 ···································································· 1

§1.2 论文研究的内容 ···································································· 2

§1.3 论文研究的意义 ···································································· 2

§1.4 论文研究的方法和技术思路以及创新点 ······································ 2

第二章数据挖掘技术研究 ····································································4

§2.1 数据挖掘的定义 ···································································· 4

§2.2 数据挖掘的研究内容与本质 ····················································· 4

§2.2.1 广义知识 ········································································5

§2.2.2 关联知识 ········································································5

§2.2.3 分类知识 ········································································5

§2.2.4 预测型知识 ·····································································6

§2.2.5 偏差型知识 ·····································································6

§2.3 数据挖掘的良性循环 ······························································ 6

§2.3.1 识别商业机会 ··································································7

§2.3.2 数据挖掘 ········································································8

§2.3.3 采取行动 ········································································8

§2.3.4 测试结果 ········································································8

§2.4 数据挖掘常用技术 ································································· 9

§2.4.1 分类 ············································································ 10

§2.4.2 聚类 ············································································ 10

§2.4.3 关联规则 ······································································ 11

§2.4.4 决策树 ········································································· 11

§2.4.5 人工神经网络 ································································ 12

§2.6 数据挖掘的国内外研究现状 ····················································12

§2.6.1 国外研究及应用现状 ······················································· 12

§2.6.2 国内研究及应用现状 ······················································· 13

第三章聚类分析技术及其在某茶叶有限公司应用研究 ······························14

§3.1 聚类原理与算法 ··································································· 14

§3.1.1 模糊聚类的基本原理 ······················································· 14

§3.1.2 基于模糊相似关系聚类的传递闭包法 ·································· 14

§3.2 客户价值理论 ······································································ 16

§3.3 客户价值计算方法 ································································ 17

§3.3.1 基于交易数据的客户价值模型 T-CVM ·································17

§3.3.2 基于客户价值、客户持续价值和客户损失价值的计算方法 ·······18

§3.3.3 基于信息的客户价值计算方法 ··········································· 20

§3.4 茶叶公司客户价值评价指标模型 ··············································20

§3.5 确定指标权重 ······································································ 23

§3.6 客户价值指标分数的确定 ·······················································24

§3.7 客户价值聚类 ······································································ 25

§3.7.1 客户综合价值聚类 ·························································· 25

§3.7.2 客户当前价值聚类 ·························································· 26

§3.7.3 客户潜在价值聚类 ·························································· 27

第四章关联规则技术及其在某茶叶有限公司应用研究 ······························30

§4.1 关联规则的规则形式 ·····························································30

§4.1.1 支持度与置信度 ····························································· 30

§4.1.2 提升度 ········································································· 31

§4.2 关联规则算法 ······································································ 31

§4.2.2 关联规则挖掘的核心算法——Apriori ·································· 32

§4.2.2 基于关系代数理论的关联规则挖掘算法——ORAR ················ 33

§4.2.3 基于概念分层的泛化关联规则挖掘算法——RGAR ················ 35

§4.3 关联规则在某茶叶有限公司的应用 ···········································38

§4.3.1 支持度的计算 ································································ 39

§4.3.2 强规则的产生 ································································ 41

§4.3.3 提升度 ········································································· 42

§4.4 关联规则的扩展思想 ·····························································43

第五章基于数据挖掘技术的某茶叶有限公司决策支持系统分析 ················· 46

§5.1 决策支持系统的诞生及发展 ····················································46

§5.1.1 决策支持系统的诞生 ······················································· 46

§5.1.2 数据仓库的诞生及对决策支持系统的推动作用 ······················46

§5.2 基于数据挖掘的新决策支持系统 ··············································47

§5.2.1 传统智能决策支持系统的体系结构及其不足 ·························47

§5.2.2 基于数据挖掘的新决策支持系统结构体系 ····························48

§5.3 某茶叶有限公司决策支持系统设计 ···········································49

§5.3.1 背景介绍 ······································································ 49

§5.3.2 数据仓库的建立 ····························································· 50

§5.3.3 数据挖掘算法的应用 ······················································· 50

§5.3.4 某茶叶有限公司决策支持系统结构体系 ·······························51

§5.4 决策支持系统软件包 ·····························································52

第六章总结与展望 ··········································································· 54

§6.1 总结 ·················································································· 54

§6.2 进一步的研究 ······································································ 54

参考文献 ·························································································56

在读期间公开发表的论文和承担科研项目及取得成果 ······························· 60

致谢 ······························································································· 61

第一章绪论

§1.1 论文研究的背景

零售业[1]是市场经济最前沿、最活跃的行业，它必须时刻紧贴瞬息万变的市场

脉搏，与零售业相适应的信息化建设同样也是最前沿、最活跃的，而且也是最易

变化的。中国的零售业在过去的两年中，无论是外资企业还是本土企业，都得到

了迅速的发展。企业的规模扩大了，管理水平需要提高了，企业的信息化问题也

随之提到了日程上来。很多从事零售信息技术产品研发、生产的国内外公司，也

纷纷将他们的新技术、新产品和解决方案展现在中国零售业面前；也引起了中国

零售商的关注。虽然在过去的十年中，中国的零售企业在信息技术的应用上，取

得了可喜的成绩，但是我们不得不承认，大多数中国零售企业的信息化仍然处于

一个较低的水平，很多基础的东西还需要我们进一步去打造。如果这些基础事情

没有做好，再先进的技术，再好的产品都不可能发挥其应有的作用。

纵观现今中国大多数零售企业，虽然在一定程度上应用了基础的信息化，但

是以进销调存结（算）为核心的、以C/S为主要技术指标的商业计算机管理系统，

已经不适应现代商业发展的需要。而以数据集中、三层或多层或B/S架构、采用大

型数据库系统、商品连锁配送（物流）、经营业态的纵向系统专业化、横向的综合

性系统走集成化等方向发展。同时随着零售企业的迅速发展以及数据库系统的广

泛应用，许多零售企业积累了大量的数据，这些海量数据收集、存放在大型数据

库中。由于缺乏有效的工具，理解它们已经远远超出了人们的能力。结果，这些

数据成了难得再访问的数据档案。决策者所做出的决策往往不是基于数据库中信

息丰富的数据，而是基于决策者的直觉。人们迫切需要强有力的工具来挖掘数据

背后隐藏的知识，应用于决策管理、生产控制、市场分析、科学探索等领域。虽

然目前数据库系统已日趋成熟，并且也包含了一些联机分析处理(OLAP)工具，具

有汇总、合并和聚集等功能，以及从不同的角度观察信息的能力，但它仍只是一

种验证式分析工具，不能解决深层次的分析，如数据分类、聚类和数据随时间变

化的特征，因此仍然需要更有效的分析工具。数据挖掘技术正是在这样的应用需

求推动下产生并迅速发展起来的。数据挖掘的出现为自动和智能地把海量的数据

转化为有用的信息和知识提供了手段。它把人们对数据的应用从低层次的简单查

询，提升到从数据中挖掘知识，提供决策支持。

本人在某茶叶有限公司实习期间一直从事企业信息化建设和数据挖掘、数据

分析方面的工作，深刻体会到该公司对企业管理信息化以及对各种历史数据进行

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

15 积分 4人已下载

立即下载 VIP免费下载

摘要：

摘要近年来，随着科学技术的飞速发展，经济和社会都取得了极大的进步，与此同时，在各个领域产生了大量的数据，激增的数据背后隐藏着许多重要的信息。人们不再满足于数据库的查询功能，希望能够对其进行更高层次的分析，以便能从数据中提取信息和知识为决策服务。数据挖掘的出现为自动和智能地把海量的数据转化为有用的信息和知识提供了手段，它把人们对数据的应用从低层次的简单查询，提升到从数据中挖掘知识，提供决策支持。本文首先以某茶叶有限公司为依托，深入分析数据挖掘中的模糊聚类与关联规则技术，重点讨论基于模糊相似关系聚类的传递闭包法、基于关系代数理论的关联规则挖掘算法——ORAR、以及基于概念分层的泛化关联规则挖掘算法...

展开>> 收起<<

数据挖掘在零售企业信息化建设中的应用研究.pdf

共64页,预览7页

还剩页未读，继续阅读

数据挖掘在零售企业信息化建设中的应用研究

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

推荐作者

热门标签

举报选择: