CRM系統(tǒng)：基于OLAM技術(shù)的分析型CRM設(shè)計(jì)與開(kāi)發(fā)_英文_

3　Softw a re Realization Process of the A nalytica l CRM System B ased on DM The softw are p rocess of the analytical CRM system based on OLAM is as follow s: 　　Fig. 3 　Customer analysis and m ining functions based on OLAM technology 1) Selection of softw are and hardw are. M ain soft2 w ares:W indow s 2003 Server enterp rise edition (sql, in2 cluding IIS); SQL Server 2005 simp le Chinese, corpora2 tion edition. M ain hardw ares: one server (Rackmount server) , w hich is suitable for develop ing DB server / data analysis server in LAN environments and used to build the data w arehouse and the data m ining m odel; the other app lication2server (desktop ) is to be a server of the oper2 ational system, w hich is used as a server in develop ing environments. 2) System architecture. The B /S model is adop ted. Operational systemps developm ent uses SQL Server 2005 DBMS to realize the control and management of the source code. 3) D ata p rep rocessing. D ata p rep rocessing is one important step in the data m ining (know ledge discovery) p rocess. Especially for the data m ining of noise data, in2 comp lete data, and inconsistent data, it imp roves the quality of the objects of data m ining and reaches the pur2 pose of imp roving models and know ledge obtained from data m ining. A fter the logic design of every top ic data w arehouse, data w arehouses w ill be built in the SQL Server 2005. 4) The app lication analysis of OLAM technology and the developm ent of analytical CRM. W hen users in2 put the parameter values of the chosen algorithms, they can obtain the analytical results of custom er classifications, the calculation of customer loyalty, and the characteris2 tics of customersp hobbies, w hich can p rovide a scientific basis for develop ing marketing strategies. Introducing fuzzy theory, analytical parameters of customer classifications are researched by com bining qualitative and quantita2 tive methods. ① The significance factors are researched w hich influence customer lifetime value based on allotting expenses. A ccording to the non2back effect of customer2w allet2shareps changes, the model of a quantitative analysis is p ro2 posed w hich can calculate exactly customer lifetime value w ith a custom er transferring matrix based on theM arkov chain. ② Parameter system is built to p redict customer loyalty degree. Combining the fuzzy neural netw ork algorithm and attribute significance theory, customer loyalty degrees can be p redicted by the fuzzy neural netw ork algorithm w ith initial param eter w eights. The effect is demonstrated better in experiments. ③ Fuzzy comp rehensive evaluation theory is studied to evaluate the credit of individual customers or enterp rise customers. Tw o kinds of cap ital credit calculation parameter systems are built. A fter every parameter w eight w hich influences client credit has been decided by means of op tim ization selection, the degree of client credit can be calcu2 lated. The CLV /CL /CC customer classification m odel is put forw ard. The results of clustering customers by the scalability K2m eans algorithm is the data p rep rocessing of customer classification by the imp roved B ayesian algo2 rithm, w hich realizes the organic combination betw een clustering and classifying. The customer purchase sequence model is studied based on association rules. The association of p roducts can be visualized by the dependence rela2 tions netw ork, and the customer sale rules can be screened out, w hich p rovide a decision2m aking basis for bundling sales, market p rom otions and so on. OLA P technology is used to drill dow n, roll up, and slice the customer and p roduct information in multi2dimensions, and analyze the potential law and the influencing factors of sales volume, the am ount of orders, etc. The graphs of development and imp lementation of the CRM system on2line analysis are show n in Figs. 4 to 6. 4　Conclusion OLAM technology is one of the most important DBMS technologies in the information industry society. It is important to app ly OLAM technology to analytical CRM development. B oth algorithm study and analysis of the p ractice need further research. There are still some concrete p roblems to analyze deep ly. [1 ] Pu Xiaoxiang, L iu W encai. An overview of on2line analytical m ining technology and its developm ent p rospect [ J ]. J ourna l of Chongqing University:Na tura l Science Edition, 2004, 27 (3) : 23 56. ( in Chinese) [2 ] Han J W, Kam berM. Da ta m ining: concep ts and techniques [M ]. M organ Kaufm ann Publisher, 2001: 20 45. [3 ] Roiger R ichard J, GeatzM ichaelW. Da ta m ining 2 a tutoria l2based p rimer [M ]. Addison2W esley Publishing, 2003: 5 18. [4 ] L iu B, Hsu W,M a Y. M ining association rules w ith multip le m inimum supports [C ] / / P roc ACM2SIGKDD Int Conf Know ledge D iscovery and Da ta M ining. San Deigo, CA, 1999: 337 341. [5 ] Shih Y2Y, L iu C2Y. A m ethod for custom er lifetim e value ranking com bining the analysis hierarchy p rocess and clustering analysis [J ]. Da tabaseMa rketing &Customer Stra tegyManagement, 2003, 11 (2) : 159 172. [6 ] Zhao Huitang,M acL ennan Jam ie. P rincip les and app lica tions of da ta m ining [M ]. Translated by Kuang Zhufang, J iao Xianlong, Gao Sheng. B eijing: Tsinghua University Press, 2007: 23 34. [7 ] Tang B ingyong,W ang Sufen. Customer rela tionship management[M ]. B eijing: Higher Education Press, 2004. ( in Chinese) [8 ] Rud O livia Parr. P ractice of da ta m ining[M ]. JohnW iley & Sons, Inc, 2005: 43 56. [9 ] Chablo Emm a. The importance of marketing data intelligence in delivering successful CRM [J ]. DM Review, 2000 (8) : 46 57. 強(qiáng)力推薦： 天柏客戶關(guān)系管理系統(tǒng) 天柏客戶關(guān)系管理系統(tǒng)（CRM）是一款集專業(yè)性、實(shí)用性、易用性為一體的純B/S架構(gòu)的CRM系統(tǒng)，它基于以客戶為中心的協(xié)同管理思想和營(yíng)銷(xiāo)理念，圍繞客戶生命周期的整個(gè)過(guò)程，針對(duì)不同價(jià)值的客戶實(shí)施以客戶滿意為目標(biāo)的營(yíng)銷(xiāo)策略，通過(guò)企業(yè)級(jí)協(xié)同，有效的“發(fā)現(xiàn)、保持和留住客戶”，從而達(dá)到留住客戶、提高銷(xiāo)售，實(shí)現(xiàn)企業(yè)利潤(rùn)最大化的目的。通過(guò)對(duì)客戶進(jìn)行7P的深入分析，即客戶概況分析(Profiling)、客戶忠誠(chéng)度分析(Persistency)、客戶利潤(rùn)分析(Profitability)、客戶性能分析(Performance)、客戶未來(lái)分析(Prospecting)、客戶產(chǎn)品分析(Product)、客戶促銷(xiāo)分析(Promotion)以及改善與管理企業(yè)銷(xiāo)售、營(yíng)銷(xiāo)、客戶服務(wù)和支持等與客戶關(guān)系有關(guān)的業(yè)務(wù)流程并提高各個(gè)環(huán)節(jié)的自動(dòng)化程度，從而幫助企業(yè)達(dá)到縮短銷(xiāo)售周期、降低銷(xiāo)售成本、擴(kuò)大銷(xiāo)售量、增加收入與盈利、搶占更多市場(chǎng)份額、尋求新的市場(chǎng)機(jī)會(huì)和銷(xiāo)售渠道，最終從根本上提升企業(yè)的核心競(jìng)爭(zhēng)力，使得企業(yè)在當(dāng)前激烈的競(jìng)爭(zhēng)環(huán)境中立于不敗之地。 關(guān)鍵詞：CRM,CRM系統(tǒng),CRM軟件,客戶關(guān)系管理,客戶管理軟件,客戶管理系統(tǒng),客戶關(guān)系管理軟件,客戶關(guān)系管理系統(tǒng)