客戶關(guān)系管理中的數(shù)據(jù)挖掘技術(shù)【外文翻譯】

1、<p><b>　　畢業(yè)論文（設(shè)計(jì)）</b></p><p><b>　　外文翻譯</b></p><p><b>　　外文原文</b></p><p>　　Data mining techniques for customerrelationship management</p>

2、;<p>　　Chris Rygielski , Jyun-Cheng Wang , David C. Yen </p><p><b>　　Abstract</b></p><p>　　Advancements in technology have made relationship marketing a reality in recent year

3、s. Technologies such as data warehousing, data mining, and campaign management software have made customer relationship management a new area where firms can gain a competitive advantage. Particularly through data mining

4、—the extraction of hidden predictive information from large databases—organizations can identify valuable customers, predict future behaviors, and enable firms to make proactive, knowledge-driven </p><p>　　2

5、002 Elsevier Science Ltd. All rights reserved.</p><p>　　Keywords: Customer relationship management (CRM); Relationship marketing; Data mining; Neural networks;Chi-square automated interaction detection (CHAI

6、D); Privacy rights</p><p>　　1. Introduction</p><p>　　A new business culture is developing today. Within it, the economics of customer relationships are changing in fundamental ways, and companie

7、s are facing the need to implement new solutions and strategies that address these changes. The concepts of mass production and mass marketing, first created during the Industrial Revolution, are being supplanted by new

8、ideas in which customer relationships are the central business issue. Firms today are concerned with increasing customer value through anal</p><p>　　The advent of the Internet has undoubtedly contributed to

9、the shift of marketing focus. As on-line information becomes more accessible and abundant, consumers become more informed and sophisticated. They are aware of all that is being offered, and they demand the best. To cope

10、with this condition, businesses have to distinguish their products or services in a way that avoids the undesired result of becoming mere commodities. One effective way to distinguish themselves is with systems that can

11、in</p><p>　　It may seem that CRM is applicable only for managing relationships between businesses and consumers. A closer examination reveals that it is even more crucial for business customers. In business-

12、to-business (B2B) environments, a tremendous amount of information is exchanged on a regular basis. For example, transactions are more numerous, custom contracts are more diverse, and pricing schemes are more complicated

13、. CRM helps smooth the process when various representatives of seller and buyer compa</p><p>　　This article examines the concepts of customer relationship management and one of its components, data mining. I

14、t begins with an overview of the concepts of data mining and CRM, followed by a discussion of evolution, characteristics, techniques, and applications of both concepts. Next, it integrates the two concepts and illustrate

15、s the relationship, benefits, and approaches to implementation, and the limitations of the technologies. Through two studies, we offer a closer look at two data mining </p><p>　　2.1. Definition</p>&l

16、t;p>　　“Data mining” is defined as a sophisticated data search capability that uses statistical algorithms to discover patterns and correlations in data [2]. The term is an analogy to gold or coal mining; data mining f

17、inds and extracts knowledge (“data nuggets”) buried in corporate data warehouses, or information that visitors have dropped on a website, most of which can lead to improvements in the understanding and use of the data. T

18、he data mining approach is complementary to other data analysis techn</p><p>　　Data mining discovers patterns and relationships hidden in data [3], and is actually part of a larger process called “knowledge

19、discovery” which describes the steps that must be taken to ensure meaningful results. Data mining software does not, however, eliminate the need to know the business, understand the data, or be aware of general statistic

20、al methods. Data mining does not find patterns and knowledge that can be trusted automatically without verification. Data mining helps business analysts</p><p>　　2.2. The evolution of data mining</p>

21、<p>　　Data mining techniques are the result of a long research and product development process. The origin of data mining lies with the first storage of data on computers, continues with improvements in data access,

22、 until today technology allows users to navigate through data in real time. In the evolution from business data to useful information, each step is built on the previous ones. Table 1 shows the evolutionary stages from t

23、he perspective of the user.</p><p>　　In the first stage, Data Collection, individual sites collected data used to make simple calculations such as summations or averages. Information generated at this step a

24、nswered business questions related to figures derived from data collection sites,such as total revenue or average total revenue over a period of time. Specific application programs were created for collecting data and ca

25、lculations. </p><p>　　The second step, Data Access, used databases to store data in a structured format. At this stage, company-wide policies for data collection and reporting of management information were

26、established. Because every business unit conformed to specific requirements or formats, businesses could query the information system regarding branch sales during any specified time period.</p><p>　　Once in

27、dividual figures were known, questions that probed the performance of aggregated sites could be asked. For example, regional sales for a specified period could be calculated. Thanks to multi-dimensional databases, a busi

28、ness could obtain either a global view or drill down to a particular site for comparisons with its peers (Data Navigation). Finally, on-line analytic tools provided real-time feedback and information exchange with collab

29、orating business units (Data Mining). This capability</p><p>　　Information systems can query past data up to and including the current level of business. Often businesses need to make strategic decisions or

30、implement new policies that better serve their customers. For example, grocery stores redesign their layout to promote more impulse purchasing. Telephone companies establish new price structures to entice customers into

31、placing more calls. Both tasks require an understanding of past customer consumption behavior data in order to identify patterns for mak</p><p>　　The core components of data mining technology have been devel

32、oping for decades in research areas such as statistics, artificial intelligence, and machine learning. Today, these technologies are mature, and when coupled with relational database systems and a culture of data integra

33、tion, they create a business environment that can capitalize on knowledge formerly buried within the systems.</p><p>　　2.3. Applications of data mining</p><p>　　Data mining tools take data and c

34、onstruct a representation of reality in the form of a model. The resulting model describes patterns and relationships present in the data. From a process orientation, data mining activities fall into three general catego

35、ries(see Fig. 1):</p><p>　　Discovery—the process of looking in a database to find hidden patterns without a predetermined idea or hypothesis about what the patterns may be.</p><p>　　Predictive M

36、odeling—the process of taking patterns discovered from the database and using them to predict the future.</p><p>　　Forensic Analysis—the process of applying the extracted patterns to find anomalous or unusua

37、l data elements.</p><p>　　Data mining is used to construct six types of models aimed at solving business problems: classification, regression, time series, clustering, association analysis, and sequence disc

38、overy [3]. The first two, classification and regression, are used to make predictions, while association and sequence discovery are used to describe behavior. Clustering can be used for either forecasting or description.

39、</p><p>　　Companies in various industries can gain a competitive edge by mining their expanding databases for valuable, detailed transaction information. Examples of such uses are provided below.</p>

40、<p>　　Each of the four applications below makes use of the first two activities of data mining: discovery and predictive modeling. The discovery process, while not mentioned explicitly in the examples (except in the

41、 retail description), is used to identify customer segments. This is done through conditional logic, analysis of affinities and associations, and trends and variations. Each of the application categories described below

42、 describes some sort of predictive modeling. Each business is intereste</p><p>　　2.3.1. Retail</p><p>　　Through the use of store-branded credit cards and point-of-sale systems, retailers can kee

43、p detailed records of every shopping transaction. This enables them to better understand their various customer segments. Some retail applications include [5]:</p><p>　　Performing basket analysis—Also known

44、as affinity analysis, basket analysis reveals which items customers tend to purchase together. This knowledge can improve stocking, store layout strategies, and promotions. Sales forecasting—Examining time-based patterns

45、 helps retailers make stocking decisions. If a customer purchases an item today, when are they likely to purchase a complementary item?</p><p>　　Database marketing—Retailers can develop profiles of customers

46、 with certain behaviors, for example, those who purchase designer labels clothing or those who attend sales. This information can be used to focus cost–effective promotions.</p><p>　　Merchandise planning and

47、 allocation—When retailers add new stores, they can improve merchandise planning and allocation by examining patterns in stores with similar demographic characteristics. Retailers can also use data mining to determine th

48、e ideal layout for a specific store.</p><p>　　2.3.2. Banking</p><p>　　Banks can utilize knowledge discovery for various applications, including [5]:</p><p>　　Card marketing—By ident

49、ifying customer segments, card issuers and acquirers can improve profitability with more effective acquisition and retention programs, targeted product development, and customized pricing.</p><p>　　Cardholde

50、r pricing and profitability—Card issuers can take advantage of data mining technology to price their products so as to maximize profit and minimize loss of customers. Includes risk-based pricing.</p><p>　　Fr

51、aud detection—Fraud is enormously costly. By analyzing past transactions that were later determined to be fraudulent, banks can identify patterns.</p><p>　　Predictive life-cycle management—Data mining helps

52、banks predict each customer’s lifetime value and to service each segment appropriately (for example, offering special deals and discounts).</p><p><b>　　譯文：</b></p><p>　　客戶關(guān)系管理中的數(shù)據(jù)挖掘技

53、術(shù)</p><p>　　Chris Rygielski , Jyun-Cheng Wang , David C. Yen </p><p><b>　　摘要</b></p><p>　　近年來，技術(shù)的進(jìn)步讓關(guān)系營銷成為一個(gè)現(xiàn)實(shí)。如數(shù)據(jù)倉庫，數(shù)據(jù)挖掘和一系列管理軟件等技術(shù)已經(jīng)取得了客戶關(guān)系管理的新領(lǐng)域，在那里公司可以贏得競爭優(yōu)勢。特別是通過數(shù)據(jù)挖

54、掘中從大型數(shù)據(jù)庫隱藏的預(yù)測信息的提取，企業(yè)可以識別有價(jià)值的客戶，預(yù)測客戶未來的行為，并使企業(yè)積極進(jìn)取，做出知識驅(qū)動(dòng)的決策。通過數(shù)據(jù)挖掘移動(dòng)可能超越過去的事件的分析，自動(dòng)化是適應(yīng)于未來的分析，通常用歷史為導(dǎo)向的工具，提供了諸如決策支持系統(tǒng)。數(shù)據(jù)挖掘工具回答了在過去太費(fèi)時(shí)追求的業(yè)務(wù)問題。然而，這些問題的答案使客戶關(guān)系管理成為可能。各種技術(shù)在數(shù)據(jù)挖掘軟件存在，不同類型的應(yīng)用程序都擁有自身的優(yōu)勢和挑戰(zhàn)。在神經(jīng)網(wǎng)絡(luò)和卡方自動(dòng)交互檢測（CHAID）

55、中存在一個(gè)特殊的二分法。雖然不同的方法于大量的境界數(shù)據(jù)挖掘，一些數(shù)據(jù)挖掘類型用于要完成的各項(xiàng)目標(biāo)的使用，對當(dāng)今的客戶關(guān)系管理理念來說是很必要的。 </p><p>　　2002 Elsevier科學(xué)有限公司保留所有權(quán)利。</p><p>　　關(guān)鍵詞：客戶關(guān)系管理（CRM） 關(guān)系營銷 數(shù)據(jù)挖掘 神經(jīng)網(wǎng)絡(luò) 卡方自動(dòng)交互檢測（CHAID） 私隱權(quán)</p><p><

56、b>　　1.簡介 </b></p><p>　　今天，一個(gè)新的商業(yè)文化正在發(fā)展。因此，客戶關(guān)系經(jīng)濟(jì)學(xué)在根本途徑中不斷變化，與此同時(shí)企業(yè)都面臨處理這些變化要實(shí)施新的解決方案和戰(zhàn)略的需要。大規(guī)模生產(chǎn)和大規(guī)模營銷的概念，最先是在工業(yè)革命時(shí)被創(chuàng)造，現(xiàn)在正在被新的觀念所取代，其中客戶關(guān)系是中央企業(yè)的問題。今天的企業(yè)越來越通過對客戶生命周期分析關(guān)注客戶價(jià)值。這些工具和數(shù)據(jù)倉庫技術(shù)，數(shù)據(jù)挖掘和其他客戶關(guān)系管理

57、（CRM）技術(shù)在關(guān)系營銷的概念上提供了新的商業(yè)機(jī)遇?！百u出—建造—重新設(shè)計(jì)”（以客為本的觀點(diǎn)）正在取代由“設(shè)計(jì)—建造—銷售”（以產(chǎn)品為導(dǎo)向的觀點(diǎn)）的舊模式。傳統(tǒng)大規(guī)模營銷的過程在一對一營銷的新方法上被質(zhì)疑。在傳統(tǒng)的過程中，營銷的目標(biāo)是吸引更多客戶，擴(kuò)大客戶群。不過，考慮到獲取新客戶的成本，它可以更好地進(jìn)行與現(xiàn)有客戶的業(yè)務(wù)。在這樣做時(shí)，營銷重點(diǎn)從客戶群的寬度轉(zhuǎn)移到每個(gè)客戶的深度需求。該性能指標(biāo)從市場份額到所謂的“錢包份額”變化。企業(yè)不只是

58、為了進(jìn)行交易而應(yīng)付客戶，他們把握了運(yùn)用服務(wù)體驗(yàn)銷售產(chǎn)品，并努力與每一位客戶建立長期合作關(guān)系的機(jī)會。</p><p>　　互聯(lián)網(wǎng)的出現(xiàn)，無疑有助于市場重點(diǎn)的轉(zhuǎn)變。由于網(wǎng)上信息變得更方便和豐富，消費(fèi)者變得更加明智和成熟。他們從所有正在提供的信息中知道，他們要求最好的。為了應(yīng)付這種情況，企業(yè)必須分清他們的產(chǎn)品或服務(wù)的方式，避免成為單純的商品這令人失望的結(jié)果。一個(gè)有效的方法來區(qū)分使用恰當(dāng)始終的與客戶進(jìn)行互動(dòng)的系統(tǒng)。收集客

59、戶的人口統(tǒng)計(jì)和行為的數(shù)據(jù)，使精確定位成為可能。這定位有助于在什么時(shí)候制定一個(gè)有效的推廣計(jì)劃，以滿足激烈的競爭或當(dāng)新產(chǎn)品出現(xiàn)時(shí)識別潛在客戶標(biāo)識。堅(jiān)持與客戶交互 ，意味著企業(yè)必須在一個(gè)網(wǎng)上系統(tǒng)中存儲交易記錄和反應(yīng)，這樣可用使有知道如何交互知識的工作人員利用。建立密切的客戶關(guān)系的重要性是公認(rèn)的和被客戶關(guān)系管理（CRM）呼吁的。</p><p>　　這可能好像CRM是只適用于企業(yè)和消費(fèi)者之間的管理關(guān)系。仔細(xì)觀察發(fā)現(xiàn)，這是

60、對商業(yè)客戶更加重要的。在企業(yè)對企業(yè)（B2B）的環(huán)境中，巨大的信息量在定期交換。例如，交易是多不勝數(shù)，合同定制更加多樣化，定價(jià)計(jì)劃是更為復(fù)雜。 CRM可幫助各代表賣方和買方公司溝通和協(xié)作的過程順利。定制目錄，個(gè)性化的商業(yè)門戶網(wǎng)站，并提供針對性的產(chǎn)品可以簡化采購進(jìn)程，提高兩家公司的效率。電子郵件警報(bào)和新產(chǎn)品信息在買方量身定做不同的角色可以幫助公司增加推銷的有效性。信任和權(quán)威是否有針對性的學(xué)術(shù)報(bào)告或行業(yè)消息被傳遞到有關(guān)人士的加強(qiáng)。這些全可被視

61、為在客戶關(guān)系管理的好處。凱捷安進(jìn)行了一項(xiàng)研究，以了解公司認(rèn)識和一個(gè)CRM戰(zhàn)略[1]的準(zhǔn)備。在接受調(diào)查的公司中，65％意識到CRM技術(shù)和方法，28％有對CRM項(xiàng)目有所研究或在實(shí)施階段; 12％是在實(shí)際運(yùn)作階段。 45％的受訪公司，對CRM項(xiàng)目的實(shí)施和已開展由最高控制管理層的監(jiān)測。因此，很明顯，這是一個(gè)看作是一個(gè)關(guān)鍵的戰(zhàn)略舉措的新興概念。</p><p>　　本文探討了客戶關(guān)系管理理念和其一組成部分，數(shù)據(jù)挖掘。它始于

62、一個(gè)數(shù)據(jù)挖掘和客戶關(guān)系管理的概念的概述，接著討論演化、特點(diǎn)、技術(shù)中的應(yīng)用,并闡述了兩個(gè)概念。接下來，它集成了兩個(gè)概念，并說明的關(guān)系，利益，以及實(shí)現(xiàn)方法，以及局限性的技術(shù)。經(jīng)過兩年的研究，我們提供了兩個(gè)數(shù)據(jù)挖掘細(xì)看技巧：卡方自動(dòng)交互檢測（CHAID）和神經(jīng)網(wǎng)絡(luò)。在這些案例研究的基礎(chǔ)上，CHAID和神經(jīng)網(wǎng)絡(luò)在各自的長處和短處的基礎(chǔ)上進(jìn)行了比較和對比。最后，我們根據(jù)討論得出了的結(jié)論。</p><p>　　2.數(shù)據(jù)挖掘：

63、概述 2.1.定義 </p><p>　　“數(shù)據(jù)挖掘”被定義為一個(gè)使用統(tǒng)計(jì)算法來發(fā)現(xiàn)數(shù)據(jù)模式和相關(guān)性[2]復(fù)雜的數(shù)據(jù)搜索功能。這個(gè)詞是一個(gè)黃金和煤炭開采類比;數(shù)據(jù)挖掘發(fā)現(xiàn)和提取埋葬在企業(yè)數(shù)據(jù)倉庫里的知識（“數(shù)據(jù)掘金“），或游客從網(wǎng)站上下載的信息，其中大部分可能導(dǎo)致在理解和使用的數(shù)據(jù)上有所改進(jìn)。數(shù)據(jù)挖掘方法與其他數(shù)據(jù)分析是相輔相成的，如統(tǒng)計(jì)技術(shù)，在線分析處理（OLAP），電子表格，和基礎(chǔ)數(shù)據(jù)的訪問。簡單來說，數(shù)據(jù)

64、挖掘是另一種方式在數(shù)據(jù)中找到意義。</p><p>　　數(shù)據(jù)挖掘發(fā)現(xiàn)數(shù)據(jù)中隱藏的模式和關(guān)系[3]，其實(shí)是一部分所謂的“知識發(fā)現(xiàn)”的更大過程，用來描述一個(gè)必須采取確保有意義的結(jié)果的步驟。數(shù)據(jù)挖掘軟件也不具備，但是，消除了解業(yè)務(wù)，理解數(shù)據(jù)，或者是普遍知道統(tǒng)計(jì)方法的需要。數(shù)據(jù)挖掘沒有找到模式和可信任的情況下自動(dòng)驗(yàn)證的知識。數(shù)據(jù)挖掘幫助業(yè)務(wù)分析師產(chǎn)生假設(shè)，但它不驗(yàn)證這假說。</p><p>　　2

65、.2.數(shù)據(jù)挖掘的演化 </p><p>　　數(shù)據(jù)挖掘技術(shù)是一個(gè)長期的研究和產(chǎn)品發(fā)展過程的結(jié)果。數(shù)據(jù)挖掘根源在于第一個(gè)存儲數(shù)據(jù)的計(jì)算機(jī)上，繼續(xù)與數(shù)據(jù)訪問的改進(jìn)，直到今天的技術(shù)允許用戶瀏覽實(shí)時(shí)數(shù)據(jù)。在從業(yè)務(wù)數(shù)據(jù)到有用的信息進(jìn)化中，每一步都是建立在之前的每一步。表1從用戶的角度上顯示了進(jìn)化的階段。在第一階段，數(shù)據(jù)收集，個(gè)別網(wǎng)站采用數(shù)據(jù)收集用來如求和或平均值的簡單計(jì)算。在此階段產(chǎn)生的信息回答了有關(guān)數(shù)據(jù)來源于數(shù)據(jù)采集地點(diǎn)的

66、商業(yè)問題，如總收入或在一段時(shí)間內(nèi)總收入的平均值。具體應(yīng)用方案是用來收集數(shù)據(jù)和計(jì)算的建立。</p><p>　　第二步，數(shù)據(jù)訪問，使用數(shù)據(jù)庫在一個(gè)結(jié)構(gòu)化的格式上存儲數(shù)據(jù)。在這個(gè)階段，全公司的數(shù)據(jù)收集政策和管理信息報(bào)告建立了。因?yàn)槊恳粋€(gè)業(yè)務(wù)單位遵照特定的要求或格式，企業(yè)可以在任何特定時(shí)間內(nèi)查詢信息系統(tǒng)對于分公司的銷售。一旦個(gè)別數(shù)字被知道，探討了聚合網(wǎng)站可以被請求的性能的問題。例如，在一個(gè)指定的期間內(nèi)地區(qū)銷售可以被計(jì)算

67、出來。到多維數(shù)據(jù)庫支持下，一個(gè)企業(yè)可以獲取無論是一個(gè)全球性的觀點(diǎn)或深入到與同行比較一個(gè)特定的網(wǎng)站（數(shù)據(jù)導(dǎo)航）。最后，在線分析工具提供實(shí)時(shí)反饋和與合作的業(yè)務(wù)單位的信息交流（數(shù)據(jù)挖掘）。這性能是非常有用的，在銷售代表或客戶服務(wù)人員需要在線找回客戶信息和回應(yīng)有關(guān)實(shí)時(shí)問題。</p><p><b>　　表1</b></p><p>　　數(shù)據(jù)挖掘演化的各階段</p>

68、<p>　　信息系統(tǒng)可以查詢到過去的數(shù)據(jù)更新和包括到目前的水平業(yè)務(wù)。為了更好的為客戶服務(wù)，企業(yè)往往需要作出的戰(zhàn)略決策或?qū)嵤┬抡?。例如，食品雜貨店重新設(shè)計(jì)自己的布局，以促進(jìn)更多的沖動(dòng)購買。電話公司建立新的價(jià)格結(jié)構(gòu)，以吸引更多的客戶配售電話。這兩項(xiàng)任務(wù)需要一個(gè)對過去的客戶消費(fèi)行為數(shù)據(jù)的了解，以便識別使用這些戰(zhàn)略決策和數(shù)據(jù)挖掘的模式——數(shù)據(jù)挖掘特別適合于這一目的。隨著先進(jìn)算法的應(yīng)用，數(shù)據(jù)挖掘揭露大量數(shù)據(jù)的認(rèn)識和指出可能有關(guān)系的

69、數(shù)據(jù)。數(shù)據(jù)挖掘幫助企業(yè)解答問題，如“在下個(gè)月，什么是波士頓可能發(fā)生的銷售額，為什么？“四個(gè)階段的每一階段都是革命性的，因?yàn)樗麄冊试S新的業(yè)務(wù)問題被回答得準(zhǔn)確并迅速[4]。在諸如統(tǒng)計(jì)，人工智能，機(jī)器學(xué)習(xí)等研究領(lǐng)域里，數(shù)據(jù)挖掘技術(shù)的核心組件已經(jīng)發(fā)展了幾十年。今天，這些技術(shù)已經(jīng)成熟，在與關(guān)系數(shù)據(jù)庫系統(tǒng)和數(shù)據(jù)集成的文化結(jié)合后，他們創(chuàng)造了可以在系統(tǒng)內(nèi)利用以前埋葬的知識的一個(gè)商業(yè)環(huán)境。</p><p>　　2.3.數(shù)據(jù)挖掘的應(yīng)

70、用</p><p>　　數(shù)據(jù)挖掘工具，取出數(shù)據(jù)并構(gòu)建一個(gè)現(xiàn)實(shí)為代表形式的模型。由此產(chǎn)生的模型描述模式和目前的數(shù)據(jù)關(guān)系。從一個(gè)進(jìn)程的方向看，數(shù)據(jù)挖掘活動(dòng)分為三大類 （見圖1）：</p><p>　　發(fā)現(xiàn) ——在沒有一個(gè)預(yù)定想法或什么模式假設(shè)下，在一個(gè)數(shù)據(jù)庫中找到隱藏模式的過程。</p><p>　　預(yù)測模型 ——從數(shù)據(jù)庫中發(fā)現(xiàn)取出模式并利用他們來預(yù)測未來的過程。 &l

71、t;/p><p>　　法醫(yī)分析 ——在應(yīng)用所提取的模式，找到異?；虿粚こ５臄?shù)據(jù)元素。</p><p>　　圖1. 過程取向下的數(shù)據(jù)挖掘故障。來源：信息發(fā)現(xiàn)。</p><p>　　數(shù)據(jù)挖掘是用來建造六個(gè)模型，旨在于解決業(yè)務(wù)類型問題：分類，回歸，時(shí)間序列，聚類，關(guān)聯(lián)分析，發(fā)現(xiàn)序列 [3]。前兩個(gè)，分類和回歸，是用來做出預(yù)測的，而關(guān)聯(lián)分析和發(fā)現(xiàn)序列是用來描述行為的。聚類可用于

72、任何預(yù)測或描述。 </p><p>　　在各個(gè)行業(yè)的公司可以通過挖掘擴(kuò)大有價(jià)值的數(shù)據(jù)庫和詳細(xì)的交易信息來獲得競爭優(yōu)勢。如下提供了這樣的例子。 </p><p>　　在下面四個(gè)應(yīng)用中每一個(gè)都利用了數(shù)據(jù)挖掘的前兩個(gè)活動(dòng)：發(fā)現(xiàn)和預(yù)測模型。發(fā)現(xiàn)過程，而沒有在例子（除了在零售描述）中明確提及，是用來鑒定客戶群。這是通過有條件的邏輯，密切分析和關(guān)聯(lián)分析，和趨勢和變化得出來的。每一個(gè)應(yīng)用程序類介紹了某種

73、預(yù)測模型的描述。每個(gè)企業(yè)都有興趣通過數(shù)據(jù)挖掘中獲得的知識來預(yù)測客戶行為。[5]。</p><p><b>　　2.3.1．零售 </b></p><p>　　通過對品牌商店的信用卡和銷售點(diǎn)系統(tǒng)的使用，零售商可以記下每一個(gè)購物交易的詳細(xì)記錄。這使他們能夠更好地了解他們不同的客戶群。一些零售應(yīng)用包括[5]： </p><p>　　表演籃分析 ——又

74、稱為親和分析，購物籃分析揭示了哪些項(xiàng)目的客戶傾向于在一起購買。這種知識可以提高辦貨，店面布局策略，促銷活動(dòng)。 </p><p>　　銷售預(yù)測 ——檢測需求模式，幫助零售商做出辦貨決定。如果今天一個(gè)客戶購買令了一個(gè)項(xiàng)目，那什么時(shí)候他們可能購買補(bǔ)充項(xiàng)目？ </p><p>　　數(shù)據(jù)庫營銷 ——零售商可以開展客戶明確行為的檔案。例如，那些誰購買名牌服裝或誰參加了銷售。這些信息可用于集中成本效益的

75、促銷活動(dòng)。 </p><p>　　商品規(guī)劃和分配 ——當(dāng)零售商增加新的商店，他們可以通過檢查存儲的類似人口統(tǒng)計(jì)特征模式改善商店規(guī)劃和分配。零售商還可以使用數(shù)據(jù)挖掘，以確定一個(gè)特定商鋪的理想布局。</p><p><b>　　2.3.2．銀行 </b></p><p>　　銀行可以利用不同應(yīng)用的知識發(fā)現(xiàn)，包括[5]： </p>&l

76、t;p>　　信用卡營銷 ——通過識別客戶群，發(fā)卡和收單可以用跟多有效地獲取去來提高盈利能力和挽留計(jì)劃，有針對性的產(chǎn)品開發(fā)，個(gè)性化的定價(jià)。 </p><p>　　持卡人定價(jià)和盈利能力 ——發(fā)卡機(jī)構(gòu)可以利用數(shù)據(jù)挖掘技術(shù)的優(yōu)勢用于其產(chǎn)品價(jià)格，以達(dá)到利潤最大化和客戶流失最小化。包括以風(fēng)險(xiǎn)為基礎(chǔ)的定價(jià)。 </p><p>　　欺詐偵測 ——欺詐是非常昂貴的。通過分析后來被確定為偽造的過去的交易