臺灣博碩士論文加值系統

國內證券業目前面臨開放外資競爭與金融控股法的實施，因此國際化、大型化與金融商品多元化是未來的趨勢。然而「大者恆大，小者恆小」的情況下，對小型的傳統經紀商而言，在商品單一和大型券商的環伺下，如何在夾縫中求生存?運用公司寶貴的資產-客戶基本資料與交易紀錄，做好客戶關係管理，或許可以為各型證券經紀商創造持續的競爭優勢。
傳統的市場行銷策略是去增加市場的佔有率，卻忽略了手續費的收入才是券商主要的利潤來源；高週轉率的客戶群，短線進出，雖對創造業績有很大的貢獻，但此客戶群也多屬高手續費折扣，對單靠仲介股票買賣為單一商品的經記商而言，手續費是唯一的收入來源，此客戶群對追求利潤的券商而言卻無多大幫助。
因此我們應該將焦點從市場佔有率移轉到客戶佔有率上，思考從客戶關係著手，區別客戶群間獲利性的差異與未來存續問題，找出忠誠度與貢獻度高的客戶，提供企業經理人針對不同客群行投資組合管理的策略，提高客戶的獲利性與留客率，並可在高貢獻度客戶流失前，先為券商示警，使券商可及時因應，擬定適切的策略“留住好客戶”。
目前由於大部分研究以業績為導向來區分客戶群，自然無法擬定適切的客戶關係策略。為驗證本研究所提出以客戶貢獻度與忠誠度為導向的模式，利用券商歷史的交易資料、持有部位的往來資料，先經由資料擷取、資料清洗和資料轉換後分三部份進行：
首先針對可能影響留客率及獲利貢獻的客戶統計、買賣行為及相關變數進行變數之整理及建檔工作。
第二階段根據前述資料將客戶區分為：流失且不具獲利性的客戶、流失且具獲利性的客戶、保留且不具獲利性的客戶和保留且具獲利性的客戶。
第三階段利用主成份分析轉換客戶資料，建立客戶貢獻度與忠誠度模式，並以模擬方式驗證分類結果。
在比較各模式分類結果之優劣，進而作出客戶關係管理及經營策略方面之建議。

With the deregulation of foreign investment and the implementation of financial holding codes, trends of globalization、expanding and multiple products have become the norm of the stock industry. However, under the course of “the big one getting more bigger and the small one getting even more smaller”, how can a small traditional broker survive with only a single product? Perhaps, by analyzing the precious resource, customer’s basic attributes and transaction records, to obtain a better customer relationship is the only choice for the company to remain a competition advantage.
Traditionally, the marketing strategy is to increase market share but ignoring the main income source, the transaction commission. Although short-term operating investors may increase the revenue to the company, but they also enjoy a higher discount rate of the fees charged. To a single product broker, the single source of income is the transaction charge from buying and selling stocks. Thus, the short-term investor did not profit single-product brokers, as it may seem.
Therefore, we move our focus from market share to customer’s retaining rate. We start from the aspect of better customer relationship, classify the customers by the profitability at institutional viewpoint, and then discover the customer group with high contribution and loyalty. This information should be useful for managers in managing investment portfolio strategies for a better customer profitability and retention rate, and in alerting the potential lost of customers with higher degree of contribution before it happened. That is, using our model, the brokers may impose appropriate strategies in order to retain “better customers” in time.
In our research, we extract historical transaction data from a broker’s database, after cleansing and transformation, we classify customers into four groups: lost without profitability, lost but profitable, retained without profitability and retained with profitability. Next, we calculate the first principal component scores of all entries; identify the distribution of the scores of each group and develop the discrimination function for classifying customers into a group. Finally we validated our model by a simulation study.

目次
摘要 i
Abstract iii
誌謝 v
目次 vi
表目錄 vii
圖目錄 viii
第一章 序論 1
第一節 研究背景與動機 1
第二節 研究目的 2
第二章 理論基礎 3
第一節 知識發掘 3
第二節 資料探勘 5
第三節 主成份分析 8
第四節 RFM分析模型 12
第三章 研究方法 15
第一節 研究流程 15
第二節 顧客屬性 17
第三節 分析潛在顧客群 19
第四章 實驗設計 22
第一節 實驗資料 22
第二節 軟體環境 25
第五章 實驗結果 30
第六章 結論 34
參考文獻 35

參考文獻
1.Bauer, C. “LA Direct Mail Customer Purchase Model,” Journal of Direct Marketing, 2, pp.16-24, 1988.
2.Berry, Micheal J.A., Linoff, Gordon S., “Data Mining Techniques – For Marketing ,Sales, and Customer Support,” John Wiley & Sons, Inc.,1997.
3.Bhatty, Mukarram et al. (Skinkle,Rod；Spalding, Thomas), “Redefining Customer loyalty, the Customer’s way,” Ivey Business Journal, pp. 13-17, Jan/Feb 2001.
4.Bult, J. R. and Wansbeek, T., “Optimal Selection For Direct Mail,” Marketing Science, 29, pp. 378-395,1995.
5.Chan, P. K., Fan, W., Andreas, L. P., and Stolfo, J. S., “Distributed Data Mining in Credit Card Fraud Detection,” IEEE Journal of Data Mining, pp. 67-74, November/December, 1999.
6.Craven, M. W. and Shavlik, J. W., “Using neural networks for data mining,” Future Generation Computer Systems, 13, pp. 221-229, 1997.
7.Fayyad, U., “The KDD process for extracting useful knowledge from volumes of data,” Communications of the ACM, 39, Iss.11, pp. 27-34, 1996.
8.Fayyad, U., “Data mining and knowledge discovery: Making sense out of data,” IEEE Expert, 11, Iss.5, pp. 20-25, 1996.
9.Fayyad, U., “Data mining and knowledge discovery in databases,” Communications of the ACM, 39, Iss. 11, pp. 22-25, 1996.
10.Goodman, John, ”Leveraging the Customer Database to your Competitive Advantage,” Direct Marketing, pp. 26-27, 1992.
11.Gorth, Robert, “Data Mining : Building Competitive Advantage,” prentice Hall PTR, 1999.
12.Hughes ,A.M., “Boosting response with RFM,” Marketing Tools, 5, pp. 4-10,1996.
13.Hughes ,A.M., “Quick Profits with RFM Analysis,” Database Marketing Institute, http://www.dbmarketing.com/articles/Art104.htm
14.Kleissner, C., “Data Mining for the Enterprise,” Proceedings of the 31st Annual Hawaii International Conference On System Sciences, pp. 295-304, July 1998.
15.Ha, Sung Ho and Sang Chan Park, “Application of data mining tools to hotel data mart on the Intranet for database marketing,” Expert Systems with Applications, 15(1), pp. 1-31, 1998.
16.Macus, C., “A practical yet meaningful approach to customer segmentation,” Journal of consumer marketing, 5, pp. 494-504, 1998.
17.McFadden, F.R., “Data Warehouse for EIS: Some Issues and Impacts,” Proceedings of the 29th Annual Hawaii International Conference On System Sciences, pp. 120-129,Wailea, HI, USA, January, 1996.
18.Miglautsch, John, “Thoughts on RFM Scoring,” Journal of Database Marketing, 8 (1), 2000.
19.Mulhern, Francis J., “Customer Profitability Analysis：Measure, Concentration, and Research Directions,” Journal of Interactive Marketing, 13(1), pp. 25-40, 1999.
20.Kahan, R., “Using Database Marketing Techniques to Enhance Your One-to-One Marketing Initiatives,” Journal of Consumer Marketing, 15, pp. 491-493, 1998.
21.Roberts, M. L., Expanding the Role of the Direct Marketing Database,” Journal of Direct Marketing, 6, pp. 51-60, 1992.
22.Shaw, Michael J., Chandrasekar Subramaniam, Gek Woo Tan and Michael E. Welge, “Knowledge Management and Data Mining for Marketing,” Journal of Decision Support Systems, 31, pp. 127-137, 2001.
23.Sung, H. H. and Sang, C.P., “Application of data mining tools to hotel data mart on the Internet for database marketing,” Expert Systems With Application, 15, pp. 1-31, 1998.
24.Stone, Bob, “Successful Direct Marketing Methods,” 4th ed., NTC Business Books, 1989.
25.陳麗君、蔡介元，「應用資料探勘技術於信用卡黃金級客戶之顧客關係管理」，元智大學工業工程與管理學系，碩士論文，民91年。
26.楊東麟，洪明傳，「資料探勘在資料倉儲的應用」，資訊與教育雜誌，pp.20-32，民國 90年8月。
27.楊清潭、趙景明，「應用資料探勘技術於顧客價值分析之研究」，私立東吳大學資訊科學系，碩士論文，民92年。
28.賴耐志，「應用資料探勘於市場區隔分析」，國立台北科技大學商業自動化與管理研究所，碩士論文，民91年。
29.藍中賢，「結合模糊集合理論與貝氏分類法之資料探勘技術」，私立元智大學資訊研究所，碩士論文，民89年。
30.蘇宥宇、邱志洲，「整合主成份分析法與模糊類神經網路模式在資料探勘分類技術上之應用」，國立台北科技大學商業自動化與管理研究所，碩士論文，民90年。
31.劉美如、楊蕙憶、林文修，「資料探勘技術在顧客區隔與目標行銷之應用」，輔仁大學資訊管理學系所，碩士論文，民91年。
32.Michael Eisen of Berkeley Lab for making the source code of Cluster/TreeView 2.11 available. M. J. L. de Hoon, S. Imoto, J. Nolan, and S. Miyano: Open Source Clustering Software. Bioinformatics, to develop Cluster 3.0.