Review of Data Mining Techniques for Detecting Churners in the Telecommunication Industry

Abstract

The telecommunication sector has been developed rapidly and with large amounts of data obtained as a result of increasing in the number of subscribers, modern techniques, data-based applications, and services. As well as better awareness of customer requirements and excellent quality that meets their satisfaction. This satisfaction raises rivalry between firms to maintain the quality of their services and upgrade them. These data can be helpfully extracted for analysis and used for predicting churners. Researchers around the world have conducted important research to understand the uses of Data mining (DM) that can be used to predict customers' churn. This paper supplies a review of nearly 73 recent journalistic articles starting in 2003 to introduce the different DM techniques used in many customerbased churning models. It epitomizes the present literature in the field of communications by highlighting the impact of service quality on customer satisfaction, detecting churners in the telecoms industry, in addition to the sample size used, the churn variables used and the results of various DM technologies. Eventually, the most common techniques for predicting telecommunication churning such as classification, regression analysis, and clustering are included, thus presenting a roadmap for new researchers to build new churn management models.