Customer churn prediction is one of the most important problems in customer relationship management (CRM). Its aim is to retain valuable customers to maximize the profit of a company. To predict whether a customer will be a churner or non-churner, there are a number of data mining techniques applied for churn prediction, such as artificial neural networks, decision trees, and support vector machines. This paper reviews some recent patents along with 21 related studies published from 2000 to 2009 and compares them in terms of the domain dataset used, data pre-processing and prediction techniques considered, etc. Future research issues are discussed.