政大機構典藏-National Chengchi University Institutional Repository(NCCUR):Item 140.119/113320

政大機構典藏 > 商學院 > 企業管理研究所（MBA學位學程） > 學位論文 > Item 140.119/113320

Please use this identifier to cite or link to this item: https://nccur.lib.nccu.edu.tw/handle/140.119/113320

Title:	應用資料採礦於電子商務平台銷售商品購買關聯性之研究 The application of data mining on the association of sales through the ecommerce platform
Authors:	李柏青
Contributors:	鄭宇庭李柏青
Keywords:	資料探勘關聯分析
Date:	2017
Issue Date:	2017-10-02 10:22:22 (UTC+8)
Abstract:	互聯網改變了傳統零售業的生態，人們的行為隨著資訊技術的發展更迭，零售結合電商已是趨勢，電商平台在其中扮演著重要的角色。面對激烈競爭的市場環境，在企業資源有限的情況下，利用用戶之巨量交易資料，透過資料探勘技術發掘出顧客行為並依此分群，進而找出潛藏具價值的關聯規則，協助電商平台進行用戶行銷，輔助電商平台現行單一式銷售的困境，此為本研究想要探討的主題。本研究利用個案公司全年的銷售紀錄作為分析基礎，運用RFM模型，以最近購買日期、購買頻率、購買金額三項變數，透過資料轉換及權重設定，計算用戶價值，以作為用戶分群之依據。將上述三項變數進行標準化後，依照資料探勘技術K-means方法將上萬用戶分為「流失客群」、「近期客群」、「次流失客群」、「潛力客群」、「重要發展客群」、「一般客群」等六群用戶群。最後透過關聯規則Apriori演算法，設定規則篩選準則後，依序找出各群用戶購買產品之行為規則，產出之結果經過領域知識的探討，最終訂立出適用不同用戶集群客製化交叉銷售的行銷方式，以提供個案公司行銷決策之輔助。 The main focus of this research is to discuss how e-commerce industries can utilize basic information, demographic and transactional data of users by applying data mining technique to find the substantial information. To isolate those of who with hidden or inconspicuous data connections to accomplish the efficiency of product marketing. This study make use of e-commerce transactions information for past whole year of the case study company as samples for analysis. Recency (R), Frequency (F), Monetary (M) are the three variables to be used. After variable adjusted and standardized, the Customer Lifetime Value is evaluated.. Using the above three variables as indices for separating users and by applying data mining skill to analyze K-means algorithm, we can apparently segment users into six groups- Leaving Customers, New Customers, Minor Leaving Customers, Potential Customers, vital Customers, Mass Customers. Then by using Association Rules Apriori, we can determine purchase connections of each group. Eventually through filters and domain knowledge, the customization of cross sales is finished, and the findings can provide the upper management functions of the case study company a more efficient direction to form a marketing campaign.
Reference:	一、英文文獻： Frawley, Piatetsky-Shapiro, Matheus (1992) Data Mining with Decision Trees Fayyad et al. (1996) Customer Intimacy Analytics : Leveraging Operational Data to Assess Customer Knowledge and Relationships and to Measure Their Business Impact Berger and Nasr (1998)Customer lifetime value: marketing models and applications Philip Kotler (2003) Marketing Management 二、中文文獻：謝邦昌、鄭宇庭與蘇志雄，2009，Data Mining概述以Clementine為例，中華資料採礦協會。鐘永富，2013年，運用資料探勘技術協助行銷策略制定之研究─以家庭清潔用品為例 Dimitri Maex,Paul B. Brown，2016，性感小數字：奧美廣告教父教你用數據讓業績一飛沖天翻譯本 R语言与统计分析, 作者: 汤银才, ISBN: 9787040250626。
Description:	碩士國立政治大學企業管理研究所(MBA學位學程) 104363006
Source URI:	http://thesis.lib.nccu.edu.tw/record/#G1043630061
Data Type:	thesis
Appears in Collections:	[企業管理研究所（MBA學位學程）] 學位論文

Files in This Item:

There are no files associated with this item.

社群 sharing

Loading...