臺灣博碩士論文加值系統

隨著網路世界且線上使用者規模不斷擴大的發展，線上預訂旅店的市場具有巨大潛力可發掘，由於預訂市場的規模增大，說明其逐漸受到各地線民的關注與使用。根據資料顯示，攜程網的市占率目前在中國線上預定旅店的市場中，佔有了50%以上的市場份額，相當龐大。研究攜程網這個領先著中國的綜合性線上旅行預訂服務公司，找出其使用者流失的關鍵因素，防止客戶流失引發的經營危機具有重要的意義。
本論文在研究客戶流失相關理論基礎上，首先將資料預處理，再利用現有特徵創造新的特徵，接著使用資料探勘方法中的XGBoost演算法、隨機森林(Random Forest)演算法、類神經網路(Multilayer Perceptron)演算法，建構以攜程網資料為基礎的客戶流失之模型，並對這些模型的性能進行評估，其次為根據建構的最優模型，分析出導致攜程網客戶流失的關鍵因素。最後根據分析的結果提出了減少攜程網客戶流失的建議。

With the development of the online world and the increasing scale of online users, the online booking hotel market has great potential to be discovered. Due to the increasing size of the booking market, it has gradually attracted the attention and use of local people. According to the data, the market share of Ctrip.com currently accounts for more than 50% of the market in China's online hotel bookings, which is quite large. It is of great significance to study Ctrip.com, a leading online travel booking service company in China, to identify the key factors of its user loss and prevent the business crisis caused by customer loss.
Based on the theory of customer churn, this paper first preprocesses the data and then uses the existing features to create new features. Then it uses the XGBoost algorithm in the data mining method, the random forest algorithm, and the neural network algorithm, constructing the model of customer churn based on Ctrip.com data, and evaluating the performance of these models. Secondly, according to the optimal model constructed, the key factors leading to the loss of Ctrip customers are analyzed. Finally, based on the results of the analysis, suggestions for reducing the loss of Ctrip customers are proposed.

摘要 II
ABSTRACT III
目錄 IV
表目錄 VII
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機及目的 3
1.3 研究意義 4
1.4 研究內容與流程 5
第二章 文獻探討 8
2.1 管理客戶關係文獻 8
2.2 相關資料探勘文獻 9
2.3 各階段客戶週期文獻 11
2.4 客戶流失文獻 12
2.5 多層感知器(Multi-Layer Perceptrons) 14
2.6 XGboost演算法 15
2.7 隨機森林演算法 19
2.8 信息增益 23
第三章 研究方法 25
3.1 資料透視 25
3.2 數據預處理 30
3.3 使用介紹 31
第四章 研究結果 34
4.1 隨機森林構建置模型 34
4.2 XGBoost構建置模型 35
4.3 類神經網路建構置模型 37
4.4 最終模型分析比較 38
4.5 影響客戶流失的關鍵因素 39
4.6 減少客戶流失對應措施 40
第五章 結論 43
5.1總結 43
5.2不足與建議 43
參考文獻 44

英文部分
[1]R. N. Bolton, P. K. Kannan, and M. D. Bramlett, "Implications of Loyalty Program Membership and Service Experiences for Customer Retention and Value, " Journal of the Academy of Marketing Science, Vol. 28, No. 1, pp. 95–108, 2000.
[2]B. J. Corbitt, "Trust and e-commerce:A study of consumer perceptions," Electronic Commerce Research and Applications, pp. 203-215, 2003.
[3]C. L. Carr ,"Hedonic and utilitarian motivations for online retail shopping behavior," Journal of Retailing ,Vol. 77, pp. 511-535 ,2001.
[4]K. Cao and P. J. Shao, "Customer Churn Prediction Based on SVM-RFE," Proceedings of International Seminar on Business and Information Management, pp. 306-309, 2008.
[5]Y. Freund and R. E. Schapire, "A decision-theoretic generalization of online learning and an application to boosting," Journal of Computer and System Sciences, Vol. 55, pp. 119-139, 1997.
[6]R. Hejazinia and M. Kazemi, "Prioritizing factors influencing customer churn," Interdisciplinary Journal of Contemporary Research In Business,Vol. 5, No. 12, pp. 227-236, 2014.
[7]M. R. Ismail, M. K. Awang and M. Makhtar, "A Multi-Layer Perceptron Approach for Customer Churn Prediction, " International Journal of Multimedia and Ubiquitous Engineering Vol.10, No.7, pp.213-222, 2015.
[8]Z. Q. John, "The Elements of Statistical Learning: Data Mining, Inference, and Prediction," Mathematical Intelligencer, Vol. 173, pp. 693–694, 2010.
[9]B. Jackson, "building Customer Relationships That Last:How Close CanIndustrial Markets Get To Their Customers-And For How Long？," Harvard Business Review, Vol. 63, No.5, pp. 120-128,1985.
[10]S. M. Keaveney, "Customer switching behavior in service industries:An exploratory study," Journal of Marketing, Vol.59, pp. 71-82, 1995.
[11]M. T. Kechadi and B. Buckley, "Customer churn prediction in telecommunications," Expert Systems with Applications,Vol.39, NO.1, pp. 14-25,2012.
[12]H. S. Kim and C. H. Yoon, "Determinants of Subscriber Churn and Customer Loyalty in the Korean Mobile Telephony Market," Telecommunications Policy, Vol.28, pp.751-765, 2008.
[13]J. Lee and L. Feick, "The impact of switching costs on the customers satisfaction loyaltylink:mobile phone service in France," Journal of Services Marketing, Vol.15, No.1, pp. 35-48, 2001.
[14]J. Lee, J. Lee and L. Feick, "The impact of switching costs on the customer satisfaction-loyalty link:mobile phone service in France," Journal of Services Marketing,Vol.15, pp. 35-48, 2010.
[15]C. S. Lin, G. H. Tzeng and Y. C. Chin, "Combined rough set theory and flow network graph to predict customer churn in credit card accounts," Expert Systems with Applications, Vol. 38, pp. 8-15, 2010.
[16]R. Mohanty and K. J. Rani ,"Application of Computatuional Intelligence to predict churn and non-churn of customers in Indian Telecommunication," Published in International Conference on Computational, pp. 12-14 ,2015.
[17]H. G. Mohammadi, P. E. Gaillardon and G. D. Micheli, "Efficient Statistical Parameter Selection for Nonlinear Modeling of process/Performance Variation," Vol.35 ,pp. 1995-2007, 2016.
[18]B. Masand, "An empirical study of product difference in consumers e-commerceadoption behavior," Electronic Commerce Research and Applications, pp.29-239, 2006.
[19]S. A. Neslin, "Defectioin: Improving Predictive Accuracy of Customer Churn Models", Vol. 3, pp. 23-27, 2004.
[20]D. K. Rigby and F. F. Reichheld and P. Schefter, "Avoid the Four Perils of CRM" , Harvard Business Review, 2002.
[21]S. Rosset and E. Neumann, "Integrating customer value considerations into predictive modeling," Published in Third IEEE International Conference on Data, pp. 2258-2263, 2003.
[22]D. C. Schmittlein and R. A. Peterson, "Customer Base Analysis:An Industrial Purchase Process Application," Marketing Science, Vol. 13, pp. 3-120, 1994.
[23]C. P. Wei and I. T. Chiu, "Turning telecommunications called tails to churn prediction:Adataminin Gap Proach," Expert Systems with Applications, Vol. 23, pp.103-112, 2002.
[24]M. Yildiz and S. Albayrak, "Customer Churn Prediction in Telecommunication with Rotation Forest Method," The Ninth International Conference on Advances in Databases, Knowledge, and Data Applications, pp.26-29, 2017.
中文部分
[25]王廣宇，"客戶關係管理（第3 版），" 北京清華大學出版社，pp. 108-109，2013
[26]王園， "基於客戶生命週期的證券業客戶細分實證研究，" 上海管理科學，
pp. 65-66，2013。
[27]王念濱，宋敏，裴大茗，"資料採擷與預測分析（第2 版），"北京清華大學出版社，2017。
[28]王政皓，"應用特徵選取和最佳化演算法提高類神經網路分類準確率，" 碩士論文，勤益科技大學工業工程與管理系研究所，2012。
[29]江柏儒，"應用SVM-RFE提升多類別分類準確率，"，碩士論文，國立勤益科技大學，2010。
[30]邵帥峰，"基於BP 神經網路對保險公司客戶流失進行分析和預測研究，" 蘭州大學，2016。
[31]林佳青，"中心通話行為預測分析顧客流失率-以A公司為例"，碩士論文，元智大學，2015。
[32]周潔如，莊輝，"現代客戶關係管理，" 上海交通大學出版社，pp. 12, 54，2009。
[33]邵兵家，"客戶關係管理(第二版)，" 北京清華大學出版社，pp. 129-132，2010。
[34]陳鴻雁，"非營利組織顧客流失率分析模式之研究─以YMCA台北萬華會所為例，" 碩士論文，輔仁大學應用統計學研究所，2005。
[35]黃文，王正林， "資料採擷：R 語言實戰，" 北京電子工業出版社，2014。
[36]黃彥齊，"類神經網路應用於空氣品質預測與異常偵測之研究"，碩士論文，2018。
[37]莊文仲，"多層感知等化器-使用進化演算法，" 碩士論文，國立中央大學，2000
[38]畢錕，"基於決策樹演算法的客戶流失預測系統的分析與研究"，武漢理工大學，2010。
[39]強南囡，"客戶關係管理，" 成都西南財經大學出版社，pp. 1-2, 49-50, 132，2013。
[40]趙宇，李兵，李秀，劉文煌，"基於改進支援向量機的客戶流失分析研究，"電腦集成製造系統，pp. 34-36，2007。
網址部分
[41]國家資訊化發展評價報告（2016），http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201611/t20161118_56109.htm
[42]第40 次《中國互聯網路發展狀況統計告》,http://www.cnnic.net.cn/hlwfzyj/hlwxzbg/hlwtjbg/201708/t20170803_69444.htm
[43]機器學習- 神經網路(多層感知機 Multilayer perceptron, MLP) 含倒傳遞( Backward propagation)詳細推導https://medium.com/@chih.sheng.huang821
[44]ROC 曲線圖https://estat.pixnet.net/blog/post/61795603-roc%E6%9B%B2%E7%B7%9A-%28receiver-operating-characteristic-curve%29
[45]準確率、召回率、精確率https://www.itread01.com/content/1548253628.html