- 期刊
以購物行為為基礎分析線上食品零售商顧客分群之研究
A Study of the Customer Segmentation Based on Shopping Behavior in Online Foodservice Retailer
摘要
本研究探討零售業如何運用RFM、顧客分群、關聯規則及決策樹分析的組合，更完整地瞭解顧客行為及有效顧客分群。資料來源為某家線上食品零售商的會員交易紀錄，以RFM指標對顧客進行分群，再以決策樹CART及Apriori法對顧客群進行購買關聯分析。研究發現將顧客分為四個群體，其消費特徵和RFM行為具顯著差異；關聯規則分析揭示了不同群體之間的商品購買關聯，這有助於定制化行銷活動，增強顧客保留率和預測顧客生命價值；決策樹分析提供了各群組的主要分類規則，有助於預測顧客類型。這三種分析技術的組合可以有效進行顧客分群並瞭解顧客購買行為，藉以掌握顧客變化與商品偏好。
並列摘要
This study investigates how the combination of RFM, customer segmentation, association rule analysis, and decision tree analysis can provide a more comprehensive understanding of customer behavior and effective customer segmentation in the retail industry. The data source is transaction records from a specific online food retailer's membership database. Customer segmentation is performed using RFM indicators, followed by CART and Apriori algorithm for analyzing purchase associations within customer groups.The study reveals that customers can be categorized into four distinct groups with significant differences in consumption characteristics and RFM behaviors. Association rule analysis uncovers purchase associations among these different customer groups, enabling customized marketing activities, enhancing customer retention rates, and predicting customer lifetime value. Additionally, decision tree analysis offers primary classification rules for each group, aiding in the prediction of customer types.The combination of these three analytical techniques allows for effective customer segmentation and provides insights into customer purchasing behaviors, facilitating an understanding of customer variations and product preferences.