企業所面臨的競爭與日俱增，為了在同業中保持競爭優勢，需以顧客服務導向為主要考量，故許多企業仰賴顧客關係管理之應用。因此如何分析會員的交易紀錄以及購買行為，藉此來增加企業的收益，就成為各企業亟欲發展的領域。本篇論文在探討顧客行為與其回購機率之關係，利用會員購買紀錄與其購買金額來建立顧客輪廓，套用在兩種分析會員回訪率之模型上，藉此來預測會員之回購率。 我們使用某知名企業所提供之資料作為實驗資料集，由實驗證明透過兩種模型能有效鑑別不同回購潛力之會員，進而尋找出有高回購潛力之會員，並且比較兩模型之間的優劣，提供企業的行銷人員能針對不同的需求，使用不同的模型來預測會員之回購率，擬定適當的行銷策略，增加會員回購的機率以及節省不必要的行銷成本，以達到增加企業收益之目標。 Enterprises are facing increasing competition, to meet various needs, companies tending to adopt differentiated and customer-oriented marketing strategies to gain competitive advantage. Customer Relationship Management (CRM) is one of the fastest growing business technology initiatives since the web. CRM leverages historical users’ behavior in order to dawn effort of enhancing customer satisfaction and loyalty. Thus, constructing a successful customer profile plays a critical role in CRM. In this study, we expect to predict the repurchase rates for the registered members at the specific category of a e-shop. However, customers’ preferences change over time. In order to capture the preference drifts of the members, In this paper, we use the members’ transactions data and monetary to construct the repurchase index (RI). However, we still could not accurately predict members’ repurchase rate by RI. Then, we use Behavioral Repurchase model (BRM) and Repurchase Index with Time Factor (RIT) model to effectively predict repurchase rates. We will test and verify the accuracy of two models. In the end, we compare the accuracy of two models. The marketers of the e-shop can target the registered members with high repurchase rates and design corresponding marketing strategies.