Razieh Qiasi - University of Qom
Zahra Roozbehani - University of Shahid Beheshti
Behrooz Minaei-Bidgoli - University of Science and Technology

A major concern for modern enterprises is to promote customer value, loyalty and contribution through services which can help establishing long-term relationshipswith customers. Organizations have found that retaining existing customers is more valuable than attracting new customers. Therefore, preventing customer churn by customer retention to achieve maximum profit is a critical issue in customer relationship management. In order to effectivelymanage customer churn for companies, it is important to build a more effective and accurate customer churn prediction model. Data mining and statistical techniques can be used to construct prediction models. This paper aims to identify most appropriate models base on data mining techniques. In this paper, rough set theory has been used for feature selection. It aims to find the most effective features in order to reducecustomer loss. Then, neural networks are used in order to create the model. Finally, to evaluate performance of the model five measures (accuracy, precision, Recall, F-measure, Lift) were used. Results show that our proposed model provides acceptable performance in terms of evaluation measures.

Customer relationship management, Data mining, Customer Churn, neural network, rough set theory