電腦運算功能日益增強,企業過去至今也累積龐大之數據資料,資料探勘技術隨之蓬勃發展。企業逐漸意識到透過資料探勘方式對於決策的價值,本研究取用某線上食品零售業者交易數據進行資料探勘,以提出顧客關係管理方案。 本研究根據電商4P之概念建構顧客關係管理模式,以RFM指標對線上食品零售公司的顧客進行兩階段K-means分群,形成4種具有顯著差異的顧客群體,再以決策樹CART以及Apriori法對顧客群進行資料探勘。 根據研究結果,集群分析將顧客區分為「鮮肉型顧客」、「沉睡巨人型顧客」、「忠誠型顧客」、「流失型顧客」四群,進一步透過決策樹CART與Apriori法掌握各群顧客特徵與產品購買關聯性,以期作為日後企業對顧客群廣告投放、行銷預測及服務策略擬定之參考依據。
Computer computing functions are increasing day by day. The companies has also accumulated huge amounts of data in the past. This has led to the popularization of data mining technology. Many companies are gradually realizing the value of data mining for decisions. This research used the transaction data of the online food retailer for data mining to propose a customer relationship management plan. This research is based on concept of e-commerce 4P to construct a customer relationship management model. Using RFM indicators for two-stage K-means clustering on customers of online food retail company. There were 4 types of customer groups with significant differences formed. Then use the decision tree CART and the Apriori method to mine the data of the customer groups. According to the research results, cluster analysis divides customers into 4 groups: fresh meat customers, sleeping giant customers, loyalty customers and churning customers. Further, through the decision tree CART and the Apriori method, grasp customers characteristics and product relevance. It’s expected to be used as a reference for advertising, marketing prediction and service strategy in the future.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。