【摘要】：面對不斷變化的市場需求和競爭激烈的市場環(huán)境，將客戶流失率降至最低，是企業(yè)贏得市場、取得成功的根本。自從我國加入WTO后，各個市場對外開放，，我國各行各業(yè)都面臨來自外國產(chǎn)品的競爭，大大加重了企業(yè)對客戶資源的爭奪，與此同時，信息技術(shù)的快速發(fā)展推動了電子商務(wù)時代的到來，網(wǎng)絡(luò)營銷以其獨特的優(yōu)勢運營而生，使得供應(yīng)商也成為企業(yè)的競爭對手，這對零售企業(yè)來說無疑是雪上加霜。因此，對零售行業(yè)的進行客戶關(guān)系管理顯得迫切而重要。 客戶關(guān)系管理的重點在于降低客戶流失率，而降低客戶流失率的關(guān)鍵在于客戶流失預警，進行客戶流失預警的常用技術(shù)是數(shù)據(jù)挖掘。本論文在客戶關(guān)系管理的大知識背景框架下，梳理分析客戶流失管理相關(guān)理論的基礎(chǔ)上，利用數(shù)據(jù)挖掘技術(shù)方法對零售業(yè)客戶流失預測問題進行了研究。本文首先回顧了客戶關(guān)系管理的相關(guān)理論知識，客戶流失的定義、原因及客戶流失管理過程，探討了客戶價值的概念以及客戶價值評估的幾類算法。然后提出了基于RFM客戶價值和IG-NN屬性選擇的客戶流失預警模型，利用RFM模型計算出客戶價值，用信息增益選擇主要屬性，再用神經(jīng)網(wǎng)絡(luò)分析每個主要屬性對客戶流失率的影響程度并結(jié)合二八法則判斷導致客戶流失的關(guān)鍵屬性，并以客戶價值、關(guān)鍵屬性作為神經(jīng)網(wǎng)絡(luò)的輸入，客戶流失概率作為網(wǎng)絡(luò)輸出，構(gòu)建基于RFM客戶價值和IG-NN屬性選擇的客戶流失預警模型。然后將本文所得結(jié)果與單一神經(jīng)網(wǎng)絡(luò)和基于IG-NN屬性選擇的客戶流失預警模型進行對比，發(fā)現(xiàn)本文的客戶流失預警模型在準確率、命中率、覆蓋率以及提升度方面均優(yōu)于另外兩個模型。最后，對本文的研究結(jié)論進行了總結(jié)，并對未來研究提出展望。
[Abstract]:Facing the changing market demand and the fierce competition market environment, it is the foundation for the enterprise to win the market and achieve success by reducing the customer wastage rate to the lowest. Since China's entry into WTO, various markets have been opened to the outside world, and all kinds of industries in our country are facing competition from foreign products, which greatly aggravates the competition for customer resources by enterprises. At the same time, The rapid development of information technology has promoted the arrival of the era of electronic commerce. Network marketing is born with its unique advantages, which makes suppliers become competitors of enterprises, which is undoubtedly even worse for retail enterprises. Therefore, it is urgent and important to carry out customer relationship management in retail industry. The focus of customer relationship management is to reduce the customer turnover rate, and the key to reduce the customer turnover rate lies in customer loss early warning. Data mining is the common technology to carry out customer turnover early warning. Under the framework of customer relationship management (CRM), this paper combs and analyzes the related theories of customer turnover management, and studies the prediction of retail customer turnover by using data mining technology. This paper first reviews the relevant theoretical knowledge of customer relationship management, the definition of customer loss, the causes and the process of customer loss management, and discusses the concept of customer value and several kinds of algorithms of customer value evaluation. Then a customer loss early warning model based on RFM customer value and IG-NN attribute selection is proposed. RFM model is used to calculate the customer value, and the information gain is used to select the main attributes. Then the influence degree of each main attribute on the customer turnover rate is analyzed by neural network, and the key attributes leading to customer loss are judged by the 28 ~ (th) rule, and the customer value and key attributes are taken as the input of the neural network. As the network output, the customer loss probability is used as the network output, and the customer loss early warning model based on RFM customer value and IG-NN attribute selection is constructed. Then the results obtained in this paper are compared with the single neural network and the customer turnover early warning model based on IG-NN attribute selection, and it is found that the customer turnover early warning model in this paper has the accuracy and hit rate. Coverage and improvement are superior to the other two models. Finally, the research conclusions of this paper are summarized, and the prospect of future research is put forward.
【學位授予單位】：吉林大學
【學位級別】：碩士
【學位授予年份】：2013
【分類號】：F274;F724.2;TP183

【參考文獻】

相關(guān)期刊論文 前10條

1 馮艷;;旅游業(yè)客戶流失管理方法研究[J];商業(yè)研究;2008年08期

2 王廣宇;客戶關(guān)系管理及其在我國商業(yè)銀行領(lǐng)域的應(yīng)用[J];金融論壇;2001年07期

3 齊佳音,舒華英;電信運營業(yè)客戶價值研究的緊迫性及方向探討[J];電信科學;2003年06期

4 郭明;基于決策樹的客戶流失分析[J];廣東通信技術(shù);2004年11期

5 劉紅;謝偉;;基于客戶終身價值的流失客戶研究[J];合肥工業(yè)大學學報(社會科學版);2008年06期

6 趙宇;李兵;李秀;劉文煌;任守榘;;基于改進支持向量機的客戶流失分析研究[J];計算機集成制造系統(tǒng);2007年01期

7 鄧萬宇;鄭慶華;陳琳;許學斌;;神經(jīng)網(wǎng)絡(luò)極速學習方法研究[J];計算機學報;2010年02期

8 吳志勇;吳躍;;數(shù)據(jù)挖掘在電信業(yè)中的應(yīng)用研究[J];計算機應(yīng)用;2005年S1期

9 丁紅;陳京民;;基于數(shù)據(jù)挖掘的電信業(yè)客戶流失分析[J];中國制造業(yè)信息化;2009年07期

10 夏國恩;陳云;金煒東;;電信業(yè)客戶流失戰(zhàn)略管理模型[J];科技管理研究;2006年12期

相關(guān)博士學位論文 前1條

1 董秀華;市場準入與高校專業(yè)認證制度研究[D];華東師范大學;2004年

相關(guān)碩士學位論文 前5條

1 段書勇;基于客戶價值細分的推薦方式研究[D];吉林大學;2011年

2 李凡;數(shù)據(jù)挖掘技術(shù)的研究與應(yīng)用[D];西安電子科技大學;2002年

3 宛天巍;論客戶關(guān)系管理系統(tǒng)的實現(xiàn)與建設(shè)重點[D];南京航空航天大學;2003年

4 趙莽;基于實證分析的移動客戶保持影響因素和策略研究[D];北京郵電大學;2006年

5 王麗君;CRM中客戶價值評價體系研究[D];南京理工大學;2007年

本文編號：2477194